Anda di halaman 1dari 167


1 Introduction to Software 1.1 Evolution of the Software Testing discipline

The effective functioning of modern systems depends on our ability to produce software in a cost-effective way. The term software engineering was first used at a 1968 NATO workshop in West Germany. It focused on the growing software crisis! Thus we see that the software crisis on quality, reliability, high costs etc. started way back when most of todays software testers were not even born! The attitude towards Software Testing underwent a major positive change in the recent years. In the 1950s when Machine languages were used, testing is nothing but debugging. When in the 1960s, compilers were developed, testing started to be considered a separate activity from debugging. In the 1970s when the software engineering concepts were introduced, software testing began to evolve as a technical discipline. Over the last two decades there has been an increased focus on better, faster and cost-effective software. Also there has been a growing interest in software safety, protection and security and hence an increased acceptance of testing as a technical discipline and also a career choice!. Now to answer, What is Testing? we can go by the famous definition of Myers, which says, Testing is the process of executing a program with the intent of finding errors

1.2 The Testing process and the Software Testing Life Cycle
Every testing project has to follow the waterfall model of the testing process. The waterfall model is as given below 1.Test Strategy & Planning 2.Test Design 3.Test Environment setup 4.Test Execution 5.Defect Analysis & Tracking 6.Final Reporting According to the respective projects, the scope of testing can be tailored, but the process mentioned above is common to any testing activity. Software Testing has been accepted as a separate discipline to the extent that there is a separate life cycle for the testing activity. Involving software testing in all phases of the software development life cycle has become a necessity as part of the software quality

assurance process. Right from the Requirements study till the implementation, there needs to be testing done on every phase. The V-Model of the Software Testing Life Cycle along with the Software Development Life cycle given below indicates the various phases or levels of testing.

Requirement Study High Level Design Low Level Design Unit Testing

Production Verification Testing User Acceptance Testing System Testing Integration Testing


1.3 Broad Categories of Testing

Based on the V-Model mentioned above, we see that there are two categories of testing activities that can be done on software, namely, Static Testing Dynamic Testing The kind of verification we do on the software work products before the process of compilation and creation of an executable is more of Requirement review, design review, code review, walkthrough and audits. This type of testing is called Static Testing. When we test the software by executing and comparing the actual & expected results, it is called Dynamic Testing

1.4 Widely employed Types of Testing

From the V-model, we see that are various levels or phases of testing, namely, Unit testing, Integration testing, System testing, User Acceptance testing etc. Let us see a brief definition on the widely employed types of testing. Unit Testing: The testing done to a unit or to a smallest piece of software. Done to verify if it satisfies its functional specification or its intended design structure. Integration Testing: Testing which takes place as sub elements are combined (i.e., integrated) to form higher-level elements Regression Testing: Selective re-testing of a system to verify the modification (bug fixes) have not caused unintended effects and that system still complies with its specified requirements


System Testing: Testing the software for the required specifications on the intended hardware Acceptance Testing: Formal testing conducted to determine whether or not a system satisfies its acceptance criteria, which enables a customer to determine whether to accept the system or not. Performance Testing: To evaluate the time taken or response time of the system to perform its required functions in comparison Stress Testing: To evaluate a system beyond the limits of the specified requirements or system resources (such as disk space, memory, processor utilization) to ensure the system do not break unexpectedly Load Testing: Load Testing, a subset of stress testing, verifies that a web site can handle a particular number of concurrent users while maintaining acceptable response times Alpha Testing: Testing of a software product or system conducted at the developers site by the customer Beta Testing: Testing conducted at one or more customer sites by the end user of a delivered software product system.

1.5 The Testing Techniques

To perform these types of testing, there are two widely used testing techniques. The above said testing types are performed based on the following testing techniques. Black-Box testing technique: This technique is used for testing based solely on analysis of requirements (specification, user documentation.). Also known as functional testing. White-Box testing technique: This technique us used for testing based on analysis of internal logic (design, code, etc.)(But expected results still come requirements). Also known as Structural testing. These topics will be elaborated in the coming chapters

1.6 Chapter Summary

This chapter covered the Introduction and basics of software testing mentioning about Evolution of Software Testing

The Testing process and lifecycle Broad categories of testing Widely employed Types of Testing The Testing Techniques


2 Black Box and White Box testing

2.1 Introduction
Test Design refers to understanding the sources of test cases, test coverage, how to develop and document test cases, and how to build and maintain test data. There are 2 primary methods by which tests can be designed and they are: BLACK BOX WHITE BOX

Black-box test design treats the system as a literal "black-box", so it doesn't explicitly use knowledge of the internal structure. It is usually described as focusing on testing functional requirements. Synonyms for black-box include: behavioral, functional, opaquebox, and closed-box. White-box test design allows one to peek inside the "box", and it focuses specifically on using internal knowledge of the software to guide the selection of test data. It is used to detect errors by means of execution-oriented test cases. Synonyms for white-box include: structural, glass-box and clear-box. While black-box and white-box are terms that are still in popular use, many people prefer the terms "behavioral" and "structural". Behavioral test design is slightly different from black-box test design because the use of internal knowledge isn't strictly forbidden, but it's still discouraged. In practice, it hasn't proven useful to use a single test design method. One has to use a mixture of different methods so that they aren't hindered by the limitations of a particular one. Some call this "gray-box" or "translucent-box" test design, but others wish we'd stop talking about boxes altogether!!!

2.2 Black box testing

Black Box Testing is testing without knowledge of the internal workings of the item being tested. For example, when black box testing is applied to software engineering, the tester would only know the "legal" inputs and what the expected outputs should be, but not how the program actually arrives at those outputs. It is because of this that black box testing can be considered testing with respect to the specifications, no other knowledge of the program is necessary. For this reason, the tester and the programmer can be independent of one another, avoiding programmer bias toward his own work. For this testing, test groups are often used, Though centered around the knowledge of user requirements, black box tests do not necessarily involve the participation of users. Among the most important black box tests that do not involve users are functionality testing, volume tests, stress tests, recovery testing, and benchmarks . Additionally, there are two types of black box test that involve users, i.e. field and laboratory tests. In the following the most important aspects of these black box tests will be described briefly.

-6- Black box testing - without user involvement

The so-called ``functionality testing'' is central to most testing exercises. Its primary objective is to assess whether the program does what it is supposed to do, i.e. what is specified in the requirements. There are different approaches to functionality testing. One is the testing of each program feature or function in sequence. The other is to test module by module, i.e. each function where it is called first. The objective of volume tests is to find the limitations of the software by processing a huge amount of data. A volume test can uncover problems that are related to the efficiency of a system, e.g. incorrect buffer sizes, a consumption of too much memory space, or only show that an error message would be needed telling the user that the system cannot process the given amount of data. During a stress test, the system has to process a huge amount of data or perform many function calls within a short period of time. A typical example could be to perform the same function from all workstations connected in a LAN within a short period of time (e.g. sending e-mails, or, in the NLP area, to modify a term bank via different terminals simultaneously). The aim of recovery testing is to make sure to which extent data can be recovered after a system breakdown. Does the system provide possibilities to recover all of the data or part of it? How much can be recovered and how? Is the recovered data still correct and consistent? Particularly for software that needs high reliability standards, recovery testing is very important. The notion of benchmark tests involves the testing of program efficiency. The efficiency of a piece of software strongly depends on the hardware environment and therefore benchmark tests always consider the soft/hardware combination. Whereas for most software engineers benchmark tests are concerned with the quantitative measurement of specific operations, some also consider user tests that compare the efficiency of different software systems as benchmark tests. In the context of this document, however, benchmark tests only denote operations that are independent of personal variables. Black box testing - with user involvement

For tests involving users, methodological considerations are rare in SE literature. Rather, one may find practical test reports that distinguish roughly between field and laboratory tests. In the following only a rough description of field and laboratory tests will be given. E.g. Scenario Tests. The term ``scenario'' has entered software evaluation in the early 1990s . A scenario test is a test case which aims at a realistic user background for the evaluation of software as it was defined and performed It is an instance of black box testing where the major objective is to assess the suitability of a software product for every-day routines. In short it involves putting the system into its intended use by its envisaged type of user, performing a standardised task. In field tests users are observed while using the software system at their normal working place. Apart from general usability-related aspects, field tests are particularly useful for assessing the interoperability of the software system, i.e. how the technical integration of the system works. Moreover, field tests are the only real means to elucidate problems of the organisational integration of the software system into existing procedures. Particularly in the NLP environment this problem has frequently been underestimated. A typical example of the organisational problem of implementing a translation memory is the language service of a big automobile manufacturer, where the major implementation

problem is not the technical environment, but the fact that many clients still submit their orders as print-out, that neither source texts nor target texts are properly organised and stored and, last but not least, individual translators are not too motivated to change their working habits. Laboratory tests are mostly performed to assess the general usability of the system. Due to the high laboratory equipment costs laboratory tests are mostly only performed at big software houses such as IBM or Microsoft. Since laboratory tests provide testers with many technical possibilities, data collection and analysis are easier than for field tests.

2.3 Testing Strategies/Techniques

Black box testing should make use of randomly generated inputs (only a test range should be specified by the tester), to eliminate any guess work by the tester as to the methods of the function Data outside of the specified input range should be tested to check the robustness of the program Boundary cases should be tested (top and bottom of specified range) to make sure the highest and lowest allowable inputs produce proper output The number zero should be tested when numerical data is to be input Stress testing should be performed (try to overload the program with inputs to see where it reaches its maximum capacity), especially with real time systems Crash testing should be performed to see what it takes to bring the system down Test monitoring tools should be used whenever possible to track which tests have already been performed and the outputs of these tests to avoid repetition and to aid in the software maintenance Other functional testing techniques include: transaction testing, syntax testing, domain testing, logic testing, and state testing. Finite state machine models can be used as a guide to design functional tests According to Beizer the following is a general order by which tests should be designed: 1. Clean tests against requirements. 2. Additional structural tests for branch coverage, as needed. 3. Additional tests for data-flow coverage as needed. 4. Domain tests not covered by the above. 5. Special techniques as appropriate--syntax, loop, state, etc. 6. Any dirty tests not covered by the above.

2.4 Black box testing Methods

2.4.1 Graph-based Testing Methods
Black-box methods based on the nature of the relationships (links) among the program objects (nodes), test cases are designed to traverse the entire graph


Transaction flow testing (nodes represent steps in some transaction and links represent logical connections between steps that need to be validated) Finite state modeling (nodes represent user observable states of the software and links represent transitions between states) Data flow modeling (nodes are data objects and links are transformations from one data object to another) Timing modeling (nodes are program objects and links are sequential connections between these objects, link weights are required execution times) Equivalence Partitioning Black-box technique that divides the input domain into classes of data from which test cases can be derived An ideal test case uncovers a class of errors that might require many arbitrary test cases to be executed before a general error is observed Equivalence class guidelines: 1. If input condition specifies a range, one valid and two invalid equivalence classes are defined 2. If an input condition requires a specific value, one valid and two invalid equivalence classes are defined 3. If an input condition specifies a member of a set, one valid and one invalid equivalence class is defined 4. If an input condition is Boolean, one valid and one invalid equivalence class is defined

2.4.3 Boundary Value Analysis

Black-box technique that focuses on the boundaries of the input domain rather than its center BVA guidelines: 1. If input condition specifies a range bounded by values a and b, test cases should include a and b, values just above and just below a and b 2. If an input condition specifies and number of values, test cases should be exercise the minimum and maximum numbers, as well as values just above and just below the minimum and maximum values 3. Apply guidelines 1 and 2 to output conditions, test cases should be designed to produce the minimum and maxim output reports 4. If internal program data structures have boundaries (e.g. size limitations), be certain to test the boundaries

2.4.4 Comparison Testing

Black-box testing for safety critical systems in which independently developed implementations of redundant systems are tested for conformance to specifications Often equivalence class partitioning is used to develop a common set of test cases for each implementation

2.4.5 Orthogonal Array Testing

Black-box technique that enables the design of a reasonably small set of test cases that provide maximum test coverage Focus is on categories of faulty logic likely to be present in the software component (without examining the code) Priorities for assessing tests using an orthogonal array 1. Detect and isolate all single mode faults 2. Detect all double mode faults 3. Multimode faults

2.4.6 Specialized Testing

Graphical user interfaces Client/server architectures Documentation and help facilities Real-time systems 1. 2. 3. 4. Task testing (test each time dependent task independently) Behavioral testing (simulate system response to external events) Intertask testing (check communications errors among tasks) System testing (check interaction of integrated system software and hardware)

2.4.7 Advantages of Black Box Testing

More effective on larger units of code than glass box testing Tester needs no knowledge of implementation, including specific programming languages Tester and programmer are independent of each other Tests are done from a user's point of view Will help to expose any ambiguities or inconsistencies in the specifications Test cases can be designed as soon as the specifications are complete

2.4.8 Disadvantages of Black Box Testing

Only a small number of possible inputs can actually be tested, to test every possible input stream would take nearly forever Without clear and concise specifications, test cases are hard to design There may be unnecessary repetition of test inputs if the tester is not informed of test cases the programmer has already tried May leave many program paths untested Cannot be directed toward specific segments of code which may be very complex (and therefore more error prone) Most testing related research has been directed toward glass box testing

2.5 Black Box (Vs) White Box

- 10 -

An easy way to start up a debate in a software testing forum is to ask the difference between black box and white box testing. These terms are commonly used, yet everyone seems to have a different idea of what they mean. Black box testing begins with a metaphor. Imagine youre testing an electronics system. Its housed in a black box with lights, switches, and dials on the outside. You must test it without opening it up, and you cant see beyond its surface. You have to see if it works just by flipping switches (inputs) and seeing what happens to the lights and dials (outputs). This is black box testing. Black box software testing is doing the same thing, but with software. The actual meaning of the metaphor, however, depends on how you define the boundary of the box and what kind of access the blackness is blocking. An opposite test approach would be to open up the electronics system, see how the circuits are wired, apply probes internally and maybe even disassemble parts of it. By analogy, this is called white box testing, To help understand the different ways that software testing can be divided between black box and white box techniques, consider the Five-Fold Testing System. It lays out five dimensions that can be used for examining testing: 1.People(who does the testing) 2. Coverage (what gets tested) 3. Risks (why you are testing) 4.Activities(how you are testing) 5. Evaluation (how you know youve found a bug)

Lets use this system to understand and clarify the characteristics of black box and white box testing. People: Who does the testing? Some people know how software works (developers) and others just use it (users). Accordingly, any testing by users or other non-developers is sometimes called black box testing. Developer testing is called white box testing. The distinction here is based on what the person knows or can understand. Coverage: What is tested? If we draw the box around the system as a whole, black box testing becomes another name for system testing. And testing the units inside the box becomes white box testing. This is one way to think about coverage. Another is to contrast testing that aims to cover all the requirements with testing that aims to cover all the code. These are the two most commonly used coverage criteria. Both are supported by extensive literature and commercial tools. Requirements-based testing could be called black box because it makes sure that all the customer requirements have been verified. Code-based testing is often called white box because it makes sure that all the code (the statements, paths, or decisions) is exercised.
- 11 -

Risks: Why are you testing? Sometimes testing is targeted at particular risks. Boundary testing and other attack-based techniques are targeted at common coding errors. Effective security testing also requires a detailed understanding of the code and the system architecture. Thus, these techniques might be classified as white box. Another set of risks concerns whether the software will actually provide value to users. Usability testing focuses on this risk, and could be termed black box. Activities: How do you test? A common distinction is made between behavioral test design, which defines tests based on functional requirements, and structural test design, which defines tests based on the code itself. These are two design approaches. Since behavioral testing is based on external functional definition, it is often called black box, while structural testingbased on the code internalsis called white box. Indeed, this is probably the most commonly cited definition for black box and white box testing. Another activity-based distinction contrasts dynamic test execution with formal code inspection. In this case, the metaphor maps test execution (dynamic testing) with black box testing, and maps code inspection (static testing) with white box testing. We could also focus on the tools used. Some tool vendors refer to code-coverage tools as white box tools, and tools that facilitate applying inputs and capturing inputsmost notably GUI capture replay toolsas black box tools. Testing is then categorized based on the types of tools used. Evaluation: How do you know if youve found a bug? There are certain kinds of software faults that dont always lead to obvious failures. They may be masked by fault tolerance or simply luck. Memory leaks and wild pointers are examples. Certain test techniques seek to make these kinds of problems more visible. Related techniques capture code history and stack information when faults occur, helping with diagnosis. Assertions are another technique for helping to make problems more visible. All of these techniques could be considered white box test techniques, since they use code instrumentation to make the internal workings of the software more visible. These contrast with black box techniques that simply look at the official outputs of a program. White box testing is concerned only with testing the software product, it cannot guarantee that the complete specification has been implemented. Black box testing is concerned only with testing the specification, it cannot guarantee that all parts of the implementation have been tested. Thus black box testing is testing against the specification and will discover faults of omission, indicating that part of the specification has not been fulfilled. White box testing is testing against the implementation and will discover faults of commission, indicating that part of the implementation is faulty. In order to fully test a software product both black and white box testing are required. White box testing is much more expensive than black box testing. It requires the source code to be produced before the tests can be planned and is much more laborious in the determination of suitable input data and the determination if the software is or is not correct. The advice given is to start test planning with a black box test approach as soon as the specification is available. White box planning should commence as soon as all black box tests have been successfully passed, with the production of flowgraphs and determination of paths. The paths should then be checked against the black box test plan and any additional required test runs determined and applied.
- 12 -

The consequences of test failure at this stage may be very expensive. A failure of a white box test may result in a change which requires all black box testing to be repeated and the re-determination of the white box paths To conclude, apart from the above described analytical methods of both glass and black box testing, there are further constructive means to guarantee high quality software end products. Among the most important constructive means are the usage of object-oriented programming tools, the integration of CASE tools, rapid prototyping, and last but not least the involvement of users in both software development and testing procedures Summary : Black box testing can sometimes describe user-based testing (people); system or requirements-based testing (coverage); usability testing (risk); or behavioral testing or capture replay automation (activities). White box testing, on the other hand, can sometimes describe developer-based testing (people); unit or code-coverage testing (coverage); boundary or security testing (risks); structural testing, inspection or codecoverage automation (activities); or testing based on probes, assertions, and logs (evaluation).


Software testing approaches that examine the program structure and derive test data from the program logic. Structural testing is sometimes referred to as clear-box testing since white boxes are considered opaque and do not really permit visibility into the code. Synonyms for white box testing Glass Box testing Structural testing Clear Box testing Open Box Testing

Types of White Box testing A typical rollout of a product is shown in figure 1 below.

- 13 -

The purpose of white box testing Initiate a strategic initiative to build quality throughout the life cycle of a software product or service. Provide a complementary function to black box testing. Perform complete coverage at the component level. Improve quality by optimizing performance. Practices : This section outlines some of the general practices comprising white-box testing process. In general, white-box testing practices have the following considerations: 1. The allocation of resources to perform class and method analysis and to document and review the same. 2. Developing a test harness made up of stubs, drivers and test object libraries. 3. Development and use of standard procedures, naming conventions and libraries. 4. Establishment and maintenance of regression test suites and procedures. 5. Allocation of resources to design, document and manage a test history library. 6. The means to develop or acquire tool support for automation of capture/replay/compare, test suite execution, results verification and documentation capabilities.

1 Code Coverage Analysis

1.1 Basis Path Testing A testing mechanism proposed by McCabe whose aim is to derive a logical complexity measure of a procedural design and use this as a guide for defining a basic set of execution paths. These are test cases that exercise basic set will execute every statement at least once.

1.1.1 Flow Graph Notation

A notation for representing control flow similar to flow charts and UML activity diagrams. 1.1.2 Cyclomatic Complexity
- 14 -

The cyclomatic complexity gives a quantitative measure of 4the logical complexity. This value gives the number of independent paths in the basis set, and an upper bound for the number of tests to ensure that each statement is executed at least once. An independent path is any path through a program that introduces at least one new set of processing statements or a new condition (i.e., a new edge). Cyclomatic complexity provides upper bound for number of tests required to guarantee coverage of all program statements.

1.2 Control Structure testing

1.2.1 Conditions Testing Condition testing aims to exercise all logical conditions in a program module. They may define: Relational expression: (E1 op E2), where E1 and E2 are arithmetic expressions. Simple condition: Boolean variable or relational expression, possibly proceeded by a NOT operator. Compound condition: composed of two or more simple conditions, Boolean operators and parentheses. Boolean expression : Condition without Relational expressions. 1.2.2 Data Flow Testing Selects test paths according to the location of definitions and use of variables. 1.2.3 Loop Testing Loops fundamental to many algorithms. Can define loops as simple, concatenated, nested, and unstructured. Examples:

Note that unstructured loops are not to be tested . rather, they are redesigned. 2 Design by Contract (DbC) DbC is a formal way of using comments to incorporate specification information into the code itself. Basically, the code specification is expressed unambiguously using a formal
- 15 -

language that describes the code's implicit contracts. These contracts specify such requirements as: Conditions that the client must meet before a method is invoked. Conditions that a method must meet after it executes. Assertions that a method must satisfy at specific points of its execution Tools that check DbC contracts at runtime such as JContract [] are used to perform this function. 3 Profiling Profiling provides a framework for analyzing Java code performance for speed and heap memory use. It identifies routines that are consuming the majority of the CPU time so that problems may be tracked down to improve performance. These include the use of Microsoft Java Profiler API and Suns profiling tools that are bundled with the JDK. Third party tools such as JaViz [] may also be used to perform this function.

4 Error Handling Exception and error handling is checked thoroughly are simulating partial and complete fail-over by operating on error causing test vectors. Proper error recovery, notification and logging are checked against references to validate program design. 5 Transactions Systems that employ transaction, local or distributed, may be validated to ensure that ACID (Atomicity, Consistency, Isolation, Durability). Each of the individual parameters is tested individually against a reference data set. Transactions are checked thoroughly for partial/complete commits and rollbacks encompassing databases and other XA compliant transaction processors. Advantages of White Box Testing Forces test developer to reason carefully about implementation Approximate the partitioning done by execution equivalence Reveals errors in "hidden" code Beneficent side-effects

Disadvantages of White Box Testing Expensive Cases omitted in the code could be missed out.
- 16 -

- 17 -

GUI Testing

What is GUI Testing? GUI is the abbreviation for Graphic User Interface. It is absolutely essential that any application has to be user-friendly. The end user should be comfortable while using all the components on screen and the components should also perform their functionality with utmost clarity. Hence it becomes very essential to test the GUI components of any application. GUI Testing can refer to just ensuring that the look-and-feel of the application is acceptable to the user, or it can refer to testing the functionality of each and every component involved. The following is a set of guidelines to ensure effective GUI Testing and can be used even as a checklist while testing a product / application.

3.1 Section 1 - Windows Compliance Testing

3.1.1 Application
Start Application by Double Clicking on its ICON. The Loading message should show the application name, version number, and a bigger pictorial representation of the icon. No Login is necessary. The main window of the application should have the same caption as the caption of the icon in Program Manager. Closing the application should result in an "Are you Sure" message box Attempt to start application twice. This should not be allowed - you should be returned to main window. Try to start the application twice as it is loading. On each window, if the application is busy, then the hour glass should be displayed. If there is no hour glass, then some enquiry in progress message should be displayed. All screens should have a Help button (i.e.) F1 key should work the same. If Window has a Minimize Button, click it. Window should return to an icon on the bottom of the screen. This icon should correspond to the Original Icon under Program Manager. Double Click the Icon to return the Window to its original size. The window caption for every application should have the name of the application and the window name especially the error messages. These should be checked for spelling, English and clarity, especially on the top of the screen. Check does the title of the window make sense. If the screen has a Control menu, then use all un-grayed options. Check all text on window for Spelling/Tense and Grammar. Use TAB to move focus around the Window. Use SHIFT+TAB to move focus backwards. Tab order should be left to right, and Up to Down within a group box on the screen. All controls should get focus - indicated by dotted box, or cursor. Tabbing to an entry field with text in it should highlight the entire text in the field. The text in the Micro Help line should change - Check for spelling, clarity and non-updateable etc. If a field is disabled (grayed) then it should not get focus. It should not be possible to select them with either the mouse or by using TAB. Try this for every grayed control. Never updateable fields should be displayed with black text on a gray background with a black label. All text should be left justified, followed by a colon tight to it. In a field that
- 18 -

may or may not be updateable, the label text and contents changes from black to gray depending on the current status. List boxes are always white background with black text whether they are disabled or not. All others are gray. In general, double-clicking is not essential. In general, everything can be done using both the mouse and the keyboard. All tab buttons should have a distinct letter.

3.1.2 Text Boxes

Move the Mouse Cursor over all Enterable Text Boxes. Cursor should change from arrow to Insert Bar. If it doesn't then the text in the box should be gray or non-updateable. Refer to previous page. Enter text into Box Try to overflow the text by typing to many characters - should be stopped Check the field width with capitals W. Enter invalid characters Letters in amount fields, try strange characters like + , - * etc. in All fields. SHIFT and Arrow should Select Characters. Selection should also be possible with mouse. Double Click should select all text in box.

3.1.3 Option (Radio Buttons)

Left and Right arrows should move 'ON' Selection. So should Up and Down. Select with mouse by clicking.

3.1.4 Check Boxes

Clicking with the mouse on the box, or on the text should SET/UNSET the box. SPACE should do the same.

3.1.5 Command Buttons

If Command Button leads to another Screen, and if the user can enter or change details on the other screen then the Text on the button should be followed by three dots. All Buttons except for OK and Cancel should have a letter Access to them. This is indicated by a letter underlined in the button text. Pressing ALT+Letter should activate the button. Make sure there is no duplication. Click each button once with the mouse - This should activate Tab to each button - Press SPACE - This should activate Tab to each button - Press RETURN - This should activate The above are VERY IMPORTANT, and should be done for EVERY command Button. Tab to another type of control (not a command button). One button on the screen should be default (indicated by a thick black border). Pressing Return in ANY no command button control should activate it. If there is a Cancel Button on the screen, then pressing <Esc> should activate it. If pressing the Command button results in uncorrectable data e.g. closing an action step, there should be a message phrased positively with Yes/No answers where Yes results in the completion of the action.

3.1.6 Drop Down List Boxes

Pressing the Arrow should give list of options. This List may be scrollable. You should not be able to type text in the box. Pressing a letter should bring you to the first item in the list with that start with that letter. Pressing Ctrl - F4 should open/drop down the list box. Spacing should be compatible with the existing windows spacing (word etc.). Items should be in alphabetical order with the exception of blank/none, which is at the top or the bottom of the list box. Drop down with the item selected should be display the list with the
- 19 -

selected item on the top. Make sure only one space appears, shouldn't have a blank line at the bottom.

3.1.7 Combo Boxes

Should allow text to be entered. Clicking Arrow should allow user to choose from list

3.1.8 List Boxes

Should allow a single selection to be chosen, by clicking with the mouse, or using the Up and Down Arrow keys. Pressing a letter should take you to the first item in the list starting with that letter. If there is a 'View' or 'Open' button besides the list box then double clicking on a line in the List Box, should act in the same way as selecting and item in the list box, then clicking the command button. Force the scroll bar to appear, make sure all the data can be seen in the box.

3.2 Section 2 - Screen Validation Checklist

3.2.1 Aesthetic Conditions:
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. Is the general screen background the correct color? Are the field prompts the correct color? Are the field backgrounds the correct color? In read-only mode, are the field prompts the correct color? In read-only mode, are the field backgrounds the correct color? Are all the screen prompts specified in the correct screen font? Is the text in all fields specified in the correct screen font? Are all the field prompts aligned perfectly on the screen? Are all the field edit boxes aligned perfectly on the screen? Are all group boxes aligned correctly on the screen? Should the screen be resizable? Should the screen be allowed to minimize? Are all the field prompts spelt correctly? Are all character or alphanumeric fields left justified? This is the default unless otherwise specified. Are all numeric fields right justified? This is the default unless otherwise specified. Is all the micro-help text spelt correctly on this screen? Is all the error message text spelt correctly on this screen? Is all user input captured in UPPER case or lowercase consistently? Where the database requires a value (other than null) then this should be defaulted into fields. The user must either enter an alternative valid value or leave the default value intact. Assure that all windows have a consistent look and feel. Assure that all dialog boxes have a consistent look and feel.

3.2.2 Validation Conditions:

1. Does a failure of validation on every field cause a sensible user error message? 2. Is the user required to fix entries, which have failed validation tests? 3. Have any fields got multiple validation rules and if so are all rules being applied?
- 20 -

4. If the user enters an invalid value and clicks on the OK button (i.e. does not TAB off the field) is the invalid entry identified and highlighted correctly with an error message? 5. Is validation consistently applied at screen level unless specifically required at field level? 6. For all numeric fields check whether negative numbers can and should be able to be entered. 7. For all numeric fields check the minimum and maximum values and also some mid-range values allowable? 8. For all character/alphanumeric fields check the field to ensure that there is a character limit specified and that this limit is exactly correct for the specified database size? 9. Do all mandatory fields require user input? 10. If any of the database columns don't allow null values then the corresponding screen fields must be mandatory. (If any field, which initially was mandatory, has become optional then check whether null values are allowed in this field.)

3.2.3 Navigation Conditions:

1. Can the screen be accessed correctly from the menu? 2. Can the screen be accessed correctly from the toolbar? 3. Can the screen be accessed correctly by double clicking on a list control on the previous screen? 4. Can all screens accessible via buttons on this screen be accessed correctly? 5. Can all screens accessible by double clicking on a list control be accessed correctly? 6. Is the screen modal? (i.e.) Is the user prevented from accessing other functions when this screen is active and is this correct? 7. Can a number of instances of this screen be opened at the same time and is this correct?

3.2.4 Usability Conditions:

1. Are all the dropdowns on this screen sorted correctly? Alphabetic sorting is the default unless otherwise specified. 2. Is all date entry required in the correct format? 3. Have all pushbuttons on the screen been given appropriate Shortcut keys? 4. Do the Shortcut keys work correctly? 5. Have the menu options that apply to your screen got fast keys associated and should they have? 6. Does the Tab Order specified on the screen go in sequence from Top Left to bottom right? This is the default unless otherwise specified. 7. Are all read-only fields avoided in the TAB sequence? 8. Are all disabled fields avoided in the TAB sequence? 9. Can the cursor be placed in the microhelp text box by clicking on the text box with the mouse? 10. Can the cursor be placed in read-only fields by clicking in the field with the mouse? 11. Is the cursor positioned in the first input field or control when the screen is opened? 12. Is there a default button specified on the screen? 13. Does the default button- work correctly? - 21

14. When an error message occurs does the focus return to the field in error when the user cancels it? 15. When the user Alt+Tab's to another application does this have any impact on the screen upon return to the application? 16. Do all the fields edit boxes indicate the number of characters they will hold by there length? e.g. a 30 character field should be a lot longer

3.2.5 Data Integrity Conditions:

1. Is the data saved when the window is closed by double clicking on the close box? 2. Check the maximum field lengths to ensure that there are no truncated characters? 3. Where the database requires a value (other than null) then this should be defaulted into fields. The user must either enter an alternative valid value or leave the default value intact. 4. Check maximum and minimum field values for numeric fields? 5. If numeric fields accept negative values can these be stored correctly on the database and does it make sense for the field to accept negative numbers? 6. If a set of radio buttons represents a fixed set of values such as A, B and C then what happens if a blank value is retrieved from the database? (In some situations rows can be created on the database by other functions, which are not screen based, and thus the required initial values can be incorrect.) 7. If a particular set of data is saved to the database check that each value gets saved fully to the database. (i.e.) Beware of truncation (of strings) and rounding of numeric values.

3.2.6 Modes (Editable Read-only) Conditions:

1. 2. 3. 4. Are the screen and field colors adjusted correctly for read-only mode? Should a read-only mode be provided for this screen? Are all fields and controls disabled in read-only mode? Can the screen be accessed from the previous screen/menu/toolbar in read-only mode? 5. Can all screens available from this screen be accessed in read-only mode? 6. Check that no validation is performed in read-only mode.

3.2.7 General Conditions:

1. 2. 3. 4. 5. 6. 7. Assure the existence of the "Help" menu. Assure that the proper commands and options are in each menu. Assure that all buttons on all tool bars have a corresponding key commands. Assure that each menu command has an alternative (hot-key) key sequence, which will invoke it where appropriate. In drop down list boxes, ensure that the names are not abbreviations / cut short In drop down list boxes, assure that the list and each entry in the list can be accessed via appropriate key / hot key combinations. Ensure that duplicate hot keys do not exist on each screen 8. Ensure the proper usage of the escape key (which is to undo any changes that have been made) and generates a caution message "Changes will be lost - Continue yes/no" Assure that the cancel button functions the same as the escape key.
- 22 -


11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38.

10. Assure that the Cancel button operates, as a Close button when changes have been made that cannot be undone. Assure that only command buttons, which are used by a particular window, or in a particular dialog box, are present. (i.e) make sure they don't work on the screen behind the current screen. When a command button is used sometimes and not at other times, assures that it is grayed out when it should not be used. Assure that OK and Cancel buttons are grouped separately from other command buttons. Assure that command button names are not abbreviations. Assure that all field labels/names are not technical labels, but rather are names meaningful to system users. Assure that command buttons are all of similar size and shape, and same font & font size. Assure that each command button can be accessed via a hot key combination. Assure that command buttons in the same window/dialog box do not have duplicate hot keys. Assure that each window/dialog box has a clearly marked default value (command button, or other object) which is invoked when the Enter key is pressed - and NOT the Cancel or Close button Assure that focus is set to an object/button, which makes sense according to the function of the window/dialog box. Assure that all option buttons (and radio buttons) names are not abbreviations. Assure that option button names are not technical labels, but rather are names meaningful to system users. If hot keys are used to access option buttons, assure that duplicate hot keys do not exist in the same window/dialog box. Assure that option box names are not abbreviations. Assure that option boxes, option buttons, and command buttons are logically grouped together in clearly demarcated areas "Group Box" Assure that the Tab key sequence, which traverses the screens, does so in a logical way. Assure consistency of mouse actions across windows. Assure that the color red is not used to highlight active objects (many individuals are red-green color blind). Assure that the user will have control of the desktop with respect to general color and highlighting (the application should not dictate the desktop background characteristics). Assure that the screen/window does not have a cluttered appearance Ctrl + F6 opens next tab within tabbed window Shift + Ctrl + F6 opens previous tab within tabbed window Tabbing will open next tab within tabbed window if on last field of current tab Tabbing will go onto the 'Continue' button if on last field of last tab within tabbed window Tabbing will go onto the next editable field in the window Banner style & size & display exact same as existing windows If 8 or less options in a list box, display all options on open of list box - should be no need to scroll Errors on continue will cause user to be returned to the tab and the focus should be on the field causing the error. (i.e the tab is opened, highlighting the field with the error on it)
- 23 -

39. Pressing continue while on the first tab of a tabbed window (assuming all fields filled correctly) will not open all the tabs. 40. On open of tab focus will be on first editable field 41. All fonts to be the same 42. Alt+F4 will close the tabbed window and return you to main screen or previous screen (as appropriate), generating "changes will be lost" message if necessary. 43. Microhelp text for every enabled field & button 44. Ensure all fields are disabled in read-only mode 45. Progress messages on load of tabbed screens 46. Return operates continue 47. If retrieve on load of tabbed window fails window should not open

3.3 Specific Field Tests

3.3.1 Date Field Checks
1. Assure that leap years are validated correctly & do not errors/miscalculations. 2. Assure that month code 00 and 13 are validated correctly & do not errors/miscalculations. 3. Assure that 00 and 13 are reported as errors. 4. Assure that day values 00 and 32 are validated correctly & do not errors/miscalculations. 5. Assure that Feb. 28, 29, 30 are validated correctly & do not cause miscalculations. 6. Assure that Feb. 30 is reported as an error. 7. Assure that century change is validated correctly & does not cause miscalculations. 8. Assure that out of cycle dates are validated correctly & do not errors/miscalculations. cause cause cause errors/ errors/ cause

3.3.2 Numeric Fields

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. Assure that lowest and highest values are handled correctly. Assure that invalid values are logged and reported. Assure that valid values are handles by the correct procedure. Assure that numeric fields with a blank in position 1 are processed or reported as an error. Assure that fields with a blank in the last position are processed or reported as an error an error. Assure that both + and - values are correctly processed. Assure that division by zero does not occur. Include value zero in all calculations. Include at least one in-range value. Include maximum and minimum range values. Include out of range values above the maximum and below the minimum. Assure that upper and lower values in ranges are handled correctly.

3.3.3 Alpha Field Checks

1. Use blank and non-blank data. 2. Include lowest and highest values. 3. Include invalid characters & symbols. - 24 -

4. Include valid characters. 5. Include data items with first position blank. 6. Include data items with last position blank.

3.4 Validation Testing - Standard Actions

3.4.1 Examples of Standard Actions - Substitute your specific commands
Add View Change Delete Continue - (i.e. continue saving changes or additions) Add View Change Delete Cancel - (i.e. abandon changes or additions) Fill each field - Valid data Fill each field - Invalid data Different Check Box / Radio Box combinations Scroll Lists / Drop Down List Boxes Help Fill Lists and Scroll Tab Tab Sequence Shift Tab

3.4.2 Shortcut keys / Hot Keys

Note: The following keys are used in some windows applications, and are included as a guide. Key F1 F2 F3 F4 No Modifier Help N/A N/A N/A
- 25 -

Shift Enter Mode N/A N/A N/A

CTRL Help N/A N/A N/A Close Document

ALT N/A N/A N/A Close / Application.

Child window. F5 F6 F7 F8 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A

Toggle extend Toggle Add N/A mode, if mode, if supported. supported. N/A N/A N/A N/A N/A Move to next open Document or Child window. (Adding SHIFT reverses the order of movement). N/A

F9 F10

N/A N/A N/A Switch to previously used application. (Holding down the ALT key displays all open applications). N/A

Toggle menu bar N/A activation. N/A

F11, F12 N/A Tab

Move to next Move to active/editable previous field. active/editable field.


Puts focus on N/A first menu command (e.g. 'File').

3.4.3 Control Shortcut Keys

Key CTRL + Z CTRL + X CTRL + C CTRL + V CTRL + N Function Undo Cut Copy Paste New
- 26 -


Open Print Save Bold* Italic* Underline*

* These shortcuts are suggested for text formatting applications, in the context for which they make sense. Applications may use other modifiers for these operations.

- 27 -

4 Regression Testing 4.1 What is regression Testing

Regression testing is the process of testing changes to computer programs to make sure that the older programming still works with the new changes. Regression testing is a normal part of the program development process. Test department coders develop code test scenarios and exercises that will test new units of code after they have been written. Before a new version of a software product is released, the old test cases are run against the new version to make sure that all the old capabilities still work. The reason they might not work because changing or adding new code to a program can easily introduce errors into code that is not intended to be changed.

The selective retesting of a software system that has been modified to ensure that any bugs have been fixed and that no other previously working functions
have failed as a result of the reparations and that newly added features have not created problems with previous versions of the software. Also referred to as verification testing

Regression testing is initiated after a programmer has attempted to fix a

recognized problem or has added source code to a program that may have inadvertently introduced errors. It is a quality control measure to ensure that the newly modified code still complies with its specified requirements and that unmodified code has not been affected by the maintenance activity.

- 28 -

4.2 Test Execution

Test Execution is the heart of the testing process. Each time your application changes, you will want to execute the relevant parts of your test plan in order to locate defects and assess quality.

4.2.1 Create Test Cycles

During this stage you decide the subset of tests from your test database you want to execute. Usually you do not run all the tests at once. At different stages of the quality assurance process, you need to execute different tests in order to address specific goals. A related group of tests is called a test cycle, and can include both manual and automated tests Example: You can create a cycle containing basic tests that run on each build of the application throughout development. You can run the cycle each time a new build is ready, to determine the application's stability before beginning more rigorous testing. Example: You can create another set of tests for a particular module in your application. This test cycle includes tests that check that module in depth. To decide which test cycles to build, refer to the testing goals you defined at the beginning of the process. Also consider issues such as the current state of the application and whether new functions have been added or modified. Following are examples of some general categories of test cycles to consider:

sanity cycle checks the entire system at a basic level (breadth, rather than depth) to see that it is functional and stable. This cycle should include basic-level tests containing mostly positive checks. normal cycle tests the system a little more in depth than the sanity cycle. This cycle can group medium-level tests, containing both positive and negative checks. advanced cycle tests both breadth and depth. This cycle can be run when more time is available for testing. The tests in the cycle cover the entire application (breadth), and also test advanced options in the application (depth). regression cycle tests maintenance builds. The goal of this type of cycle is to verify that a change to one part of the software did not break the rest of the application. A regression cycle includes sanity-level tests for testing the entire software, as well as in-depth tests for the specific area of the application that was modified.

4.2.2 Run Test Cycles (Automated & Manual Tests)

Once you have created cycles that cover your testing objectives, you begin executing the tests in the cycle. You perform manual tests using the test steps. Testing Tools executes automated tests for you. A test cycle is complete only when all tests-automatic and manual-have been run.

With Manual Test Execution you follow the instructions in the test steps of each

test. You use the application, enter input, compare the application output with the - 29 -

expected output, and log the results. For each test step you assign either pass or fail status. During Automated Test Execution you create a batch of tests and launch the entire batch at once. Testing Tools runs the tests one at a time. It then imports results, providing outcome summaries for each test.

4.2.3 Analyze Test Results

After every test run one analyze and validate the test results. And have to identify all the failed steps in the tests and to determine whether a bug has been detected, or if the expected result needs to be updated.

4.3 Change Request

4.3.1 Initiating a Change Request
A user or developer wants to suggest a modification that would improve an existing application, notices a problem with an application, or wants to recommend an enhancement. Any major or minor request is considered a problem with an application and will be entered as a change request.

4.3.2 Type of Change Request

Bug the application works incorrectly or provides incorrect information. (for example, a letter is allowed to be entered in a number field) Change a modification of the existing application. (for example, sorting the files alphabetically by the second field rather than numerically by the first field makes them easier to find) Enhancement new functionality or item added to the application. (for example, a new report, a new field, or a new button)

4.3.3 Priority for the request

Low the application works but this would make the function easier or more user friendly. High the application works, but this is necessary to perform a job. Critical the application does not work, job functions are impaired and there is no work around. This also applies to any Section 508 infraction.

4.4 Bug Tracking

Locating and repairing software bugs is an essential part of software development. Bugs can be detected and reported by engineers, testers, and end-users in all phases of the testing process. fixes and determine software release dates.
- 30 -

Information about bugs must be detailed and organized in order to schedule bug

Bug Tracking involves two main stages: reporting and tracking.

4.4.1 Report Bugs

Once you execute the manual and automated tests in a cycle, you report the bugs (or defects) that you detected. The bugs are stored in a database so that you can manage them and analyze the status of your application. When you report a bug, you record all the information necessary to reproduce and fix it. You also make sure that the QA and development personnel involved in fixing the bug are notified.

4.4.2 Track and Analyze Bugs

The lifecycle of a bug begins when it is reported and ends when it is fixed, verified, and closed.

First you report New bugs to the database, and provide all necessary information to reproduce, fix, and follow up the bug. The Quality Assurance manager or Project manager periodically reviews all New bugs and decides which should be fixed. These bugs are given the status Open and are assigned to a member of the development team. Software developers fix the Open bugs and assign them the status Fixed. QA personnel test a new build of the application. If a bug does not reoccur, it is Closed. If a bug is detected again, it is reopened.

Communication is an essential part of bug tracking; all members of the development and quality assurance team must be well informed in order to insure that bugs information is up to date and that the most important problems are addressed. The number of open or fixed bugs is a good indicator of the quality status of your application. You can use data analysis tools such as re-ports and graphs in interpret bug data.

4.5 Traceability Matrix

A traceability matrix is created by associating requirements with the products that satisfy them. Tests are associated with the requirements on which they are based and the product tested to meet the requirement. Below is a simple traceability matrix structure. There can be more things included in a traceability matrix than shown below. Traceability requires unique identifiers for each requirement and product. Numbers for products are established in a configuration management (CM) plan.

- 31 -

Traceability ensures completeness, that all lower level requirements derive from higher level requirements, and that all higher level requirements are allocated to lower level requirements. Traceability is also used in managing change and provides the basis for test planning. SAMPLE TRACEABILITY MATRIX A traceability matrix is a report from the requirements database or repository. The examples below show traceability between user and system requirements. User requirement identifiers begin with "U" and system requirements with "S."

Tracing S12 to its source makes it clear this requirement is erroneous: it must be eliminated, rewritten, or the traceability corrected.

- 32 -

In addition to traceability matrices, other reports are necessary to manage requirements. What goes into each report depends on the information needs of those receiving the report(s). Determine their information needs and document the information that will be associated with the requirements when you set up your requirements database or repository

- 33 -

5 Phases of Testing
5.1 Introduction
The Primary objective of testing effort is to determine the conformance to requirements specified in the contracted documents. The integration of this code with the internal code is the important objective. Goal is to evaluate the system as a whole, not its parts Techniques can be structural or functional. Techniques can be used in any stage that tests the system as a whole (System testing ,Acceptance Testing, Unit testing, Installation, etc.)

5.2 Types and Phases of Testing

SDLC Document Software Requirement Specification Design Document Functional Specification Design Document & Functional Specs Design Document & Functional Specs Design Document & Functional Specs Unit / System / Integration Test Case Documents Functional Specs, Performance Criteria Software Requirement Specification, Unit / System / Integration / Regression / Performance Test Case Documents QA Document Requirement Checklist Design Checklist Functional Checklist Unit Test Case Documents Integration Test Case Documents System Test Case Documents Regression Test Case Documents Performance Test Case Documents User Acceptance Test Case Documents.


Acceptance Testing


System Testing

5.3 The VModel Architecture

- 34 -

Integration Testing

Detailed Design

Unit Testing


Requirement Study

Requirement Checklist Software Requirement Specification Functional Specification Checklist Functional Specification Document Architecture Design Detailed Design Document Coding

Software Requirement Specification

Functional Specification Document Architecture Design Functional Specification Document Design Document Functional

Unit Test Case Documents Unit Test Case Document System Test Case Document Integration Test Case Document Regression Test Case Document

Specification Document Unit/Integratio n/System Test Case Documents Functional Specification Performance Document Criteria Software Requirement Regression Specification Test Case Performance Document Test Cases and Scenarios

Performance Test Cases and Scenarios User Acceptance Test Case Documents/Sce narios

- 35 -

Requirement s

Regression Round 3 Performance Testing

Requirement s Review


Regression Round 2

Specification Review

System Testing


Regression Round 1

Architectur e Review

Integration Testing

Detailed Design

Design Review Code

Unit Testing

Code Walkthrough

- 36 -

6 Integration Testing
One of the most significant aspects of a software development project is the integration strategy. Integration may be performed all at once, top-down, bottom-up, critical piece first, or by first integrating functional subsystems and then integrating the subsystems in separate phases using any of the basic strategies. In general, the larger the project, the more important the integration strategy. Very small systems are often assembled and tested in one phase. For most real systems, this is impractical for two major reasons. First, the system would fail in so many places at once that the debugging and retesting effort would be impractical Second, satisfying any white box testing criterion would be very difficult, because of the vast amount of detail separating the input data from the individual code modules. In fact, most integration testing has been traditionally limited to ``black box'' techniques. Large systems may require many integration phases, beginning with assembling modules into low-level subsystems, then assembling subsystems into larger subsystems, and finally assembling the highest level subsystems into the complete system. To be most effective, an integration testing technique should fit well with the overall integration strategy. In a multi-phase integration, testing at each phase helps detect errors early and keep the system under control. Performing only cursory testing at early integration phases and then applying a more rigorous criterion for the final stage is really just a variant of the high-risk "big bang" approach. However, performing rigorous testing of the entire software involved in each integration phase involves a lot of wasteful duplication of effort across phases. The key is to leverage the overall integration structure to allow rigorous testing at each phase while minimizing duplication of effort. It is important to understand the relationship between module testing and integration testing. In one view, modules are rigorously tested in isolation using stubs and drivers before any integration is attempted. Then, integration testing concentrates entirely on module interactions, assuming that the details within each module are accurate. At the other extreme, module and integration testing can be combined, verifying the details of each module's implementation in an integration context. Many projects compromise, combining module testing with the lowest level of subsystem integration testing, and then performing pure integration testing at higher levels. Each of these views of integration testing may be appropriate for any given project, so an integration testing method should be flexible enough to accommodate them all.

Combining module testing with bottom-up integration.

- 37 -

6.1 Generalization of module testing criteria

Module testing criteria can often be generalized in several possible ways to support integration testing. As discussed in the previous subsection, the most obvious generalization is to satisfy the module testing criterion in an integration context, in effect using the entire program as a test driver environment for each module. However, this trivial kind of generalization does not take advantage of the differences between module and integration testing. Applying it to each phase of a multi-phase integration strategy, for example, leads to an excessive amount of redundant testing. More useful generalizations adapt the module testing criterion to focus on interactions between modules rather than attempting to test all of the details of each module's implementation in an integration context. The statement coverage module testing criterion, in which each statement is required to be exercised during module testing, can be generalized to require each module call statement to be exercised during integration testing. Although the specifics of the generalization of structured testing are more detailed, the approach is the same. Since structured testing at the module level requires that all the decision logic in a module's control flow graph be tested independently, the appropriate generalization to the integration level requires that just the decision logic involved with calls to other modules be tested independently. Module design complexity Rather than testing all decision outcomes within a module independently, structured testing at the integration level focuses on the decision outcomes that are involved with module calls. The design reduction technique helps identify those decision outcomes, so that it is possible to exercise them independently during integration testing. The idea behind design reduction is to start with a module control flow graph, remove all control
- 38 -

structures that are not involved with module calls, and then use the resultant "reduced" flow graph to drive integration testing. Figure 7-2 shows a systematic set of rules for performing design reduction. Although not strictly a reduction rule, the call rule states that function call ("black dot") nodes cannot be reduced. The remaining rules work together to eliminate the parts of the flow graph that are not involved with module calls. The sequential rule eliminates sequences of non-call ("white dot") nodes. Since application of this rule removes one node and one edge from the flow graph, it leaves the cyclomatic complexity unchanged. However, it does simplify the graph so that the other rules can be applied. The repetitive rule eliminates top-test loops that are not involved with module calls. The conditional rule eliminates conditional statements that do not contain calls in their bodies. The looping rule eliminates bottom-test loops that are not involved with module calls. It is important to preserve the module's connectivity when using the looping rule, since for poorly-structured code it may be hard to distinguish the ``top'' of the loop from the ``bottom.'' For the rule to apply, there must be a path from the module entry to the top of the loop and a path from the bottom of the loop to the module exit. Since the repetitive, conditional, and looping rules each remove one edge from the flow graph, they each reduce cyclomatic complexity by one. Rules 1 through 4 are intended to be applied iteratively until none of them can be applied, at which point the design reduction is complete. By this process, even very complex logic can be eliminated as long as it does not involve any module calls.

- 39 -

Incremental integration
Hierarchical system design limits each stage of development to a manageable effort, and it is important to limit the corresponding stages of testing as well. Hierarchical design is most effective when the coupling among sibling components decreases as the component size increases, which simplifies the derivation of data sets that test interactions among components. The remainder of this section extends the integration testing techniques of structured testing to handle the general case of incremental integration, including support for hierarchical design. The key principle is to test just the interaction among components at each integration stage, avoiding redundant testing of previously integrated sub-components. To extend statement coverage to support incremental integration, it is required that all module call statements from one component into a different component be exercised at each integration stage. To form- a completely flexible "statement testing" criterion, it is - 40

required that each statement be executed during the first phase (which may be anything from single modules to the entire program), and that at each integration phase all call statements that cross the boundaries of previously integrated components are tested. Given hierarchical integration stages with good cohesive partitioning properties, this limits the testing effort to a small fraction of the effort to cover each statement of the system at each integration phase. Structured testing can be extended to cover the fully general case of incremental integration in a similar manner. The key is to perform design reduction at each integration phase using just the module call nodes that cross component boundaries, yielding component-reduced graphs, and exclude from consideration all modules that do not contain any cross-component calls. Figure 7-7 illustrates the structured testing approach to incremental integration. Modules A and C have been previously integrated, as have modules B and D. It would take three tests to integrate this system in a single phase. However, since the design predicate decision to call module D from module B has been tested in a previous phase, only two additional tests are required to complete the integration testing. Modules B and D are removed from consideration because they do not contain cross-component calls, the component module design complexity of module A is 1, and the component module design complexity of module C is 2.

- 41 -

- 42 -

7 Acceptance Testing
7.1 Introduction Acceptance Testing
In software engineering, acceptance testing is formal testing conducted to determine whether a system satisfies its acceptance criteria and thus whether the customer should accept the system. The main types of software testing are: Component. Interface. System. Acceptance. Release. Acceptance Testing checks the system against the "Requirements". It is similar to systems testing in that the whole system is checked but the important difference is the change in focus: Systems Testing checks that the system that was specified has been delivered. Acceptance Testing checks that the system delivers what was requested. The customer, and not the developer should always do acceptance testing. The customer knows what is required from the system to achieve value in the business and is the only person qualified to make that judgment. The forms of the tests may follow those in system testing, but at all times they are informed by the business needs. The test procedures that lead to formal 'acceptance' of new or changed systems. User Acceptance Testing is a critical phase of any 'systems' project and requires significant participation by the 'End Users'. To be of real use, an Acceptance Test Plan should be developed in order to plan precisely, and in detail, the means by which 'Acceptance' will be achieved. The final part of the UAT can also include a parallel run to prove the system against the current system.

7.2 Factors influencing Acceptance Testing

The User Acceptance Test Plan will vary from system to system but, in general, the testing should be planned in order to provide a realistic and adequate exposure of the system to all reasonably expected events. The testing can be based upon the User Requirements Specification to which the system should conform. As in any system though, problems will arise and it is important to have determined what will be the expected and required responses from the various parties concerned; including Users; Project Team; Vendors and possibly Consultants / Contractors. In order to agree what such responses should be, the End Users and the Project Team need to develop and agree a range of 'Severity Levels'. These levels will range from (say) 1 to 6 and will represent the relative severity, in terms of business / commercial impact, of a problem with the system, found during testing. Here is an example which has been used successfully; '1' is the most severe; and '6' has the least impact :'Show Stopper' i.e. it is impossible to continue with the testing because of the severity of this error / bug Critical Problem; testing can continue but we cannot go into production (live) with this problem
- 43 -

Major Problem; testing can continue but live this feature will cause severe disruption to business processes in live operation Medium Problem; testing can continue and the system is likely to go live with only minimal departure from agreed business processes Minor Problem ; both testing and live operations may progress. This problem should be corrected, but little or no changes to business processes are envisaged 'Cosmetic' Problem e.g. colours; fonts; pitch size However, if such features are key to the business requirements they will warrant a higher severity level. The users of the system, in consultation with the executive sponsor of the project, must then agree upon the responsibilities and required actions for each category of problem. For example, you may demand that any problems in severity level 1, receive priority response and that all testing will cease until such level 1 problems are resolved. Caution. Even where the severity levels and the responses to each have been agreed by all parties; the allocation of a problem into its appropriate severity level can be subjective and open to question. To avoid the risk of lengthy and protracted exchanges over the categorisation of problems; we strongly advised that a range of examples are agreed in advance to ensure that there are no fundamental areas of disagreement; or, or if there are, these will be known in advance and your organisation is forewarned. Finally, it is crucial to agree the Criteria for Acceptance. Because no system is entirely fault free, it must be agreed between End User and vendor, the maximum number of acceptable 'outstandings' in any particular category. Again, prior consideration of this is advisable. N.B. In some cases, users may agree to accept ('sign off') the system subject to a range of conditions. These conditions need to be analysed as they may, perhaps unintentionally, seek additional functionality which could be classified as scope creep. In any event, any and all fixes from the software developers, must be subjected to rigorous System Testing and, where appropriate Regression Testing.

7.3 Conclusion
Hence the goal of acceptance testing should verify the overall quality, correct operation, scalability, completeness, usability, portability, and robustness of the functional components supplied by the Software system.

- 44 -

8.1 Introduction to SYSTEM TESTING
For most organizations, software and system testing represents a significant element of a project's cost in terms of money and management time. Making this function more effective can deliver a range of benefits including reductions in risk, development costs and improved 'time to market' for new systems. Systems with software components and software-intensive systems are more and more complex everyday. Industry sectors such as telecom, automotive, railway, and aeronautical and space, are good examples. It is often agreed that testing is essential to manufacture reliable products. However, the validation process does not often receive the required attention. Moreover, the validation process is close to other activities such as conformance, acceptance and qualification testing. The difference between function testing and system testing is that now the focus is on the whole application and its environment . Therefore the program has to be given completely. This does not mean that now single functions of the whole program are tested, because this would be too redundant. The main goal is rather to demonstrate the discrepancies of the product from its requirements and its documentation. In other words, this again includes the question, ``Did we build the right product?'' and not just, ``Did we build the product right?'' However, system testing does not only deal with this more economical problem, it also contains some aspects that are orientated on the word ``system'' . This means that those tests should be done in the environment for which the program was designed, like a mulituser network or whetever. Even security guide lines have to be included. Once again, it is beyond doubt that this test cannot be done completely, and nevertheless, while this is one of the most incomplete test methods, it is one of the most important. A number of time-domain software reliability models attempt to predict the growth of a system's reliability during the system test phase of the development life cycle. In this paper we examine the results of applying several types of Poisson-process models to the development of a large system for which system test was performed in two parallel tracks, using different strategies for test data selection. we will test that the functionality of your systems meets with your specifications, integrating with which-ever type of development methodology you are applying. We test for errors that users are likely to make as they interact with the application as well as your applications ability to trap errors gracefully. These techniques can be applied flexibly, whether testing a financial system, e-commerce, an online casino or games testing. System Testing is more than just functional testing, however, and can, when appropriate, also encompass many other types of testing, such as: o security o load/stress o performance o browser compatibility o localisation

8.2 Need for System Testing

Effective software testing, as a part of software engineering, has been proven over the last 3 decades to deliver real business benefits including:
- 45 -

reduction of costs increased productivity reduce commercial risks

Reduce rework and support overheads More effort spent on developing new functionality and less on "bug fixing" as quality increases If it goes wrong, what is the potential impact on your commercial goals? Knowledge is power, so why take a leap of faith while your competition step forward with confidence?

These benefits are achieved as a result of some fundamental principles of testing, for example, increased independence naturally increases objectivity. Your test strategy must take into consideration the risks to your organisation, commercial and technical. You will have a personal interest in its success in which case it is only human for your objectivity to be compromised.

8.3 System Testing Techniques

Goal is to evaluate the system as a whole, not its parts Techniques can be structural or functional Techniques can be used in any stage that tests the system as a whole (acceptance, installation, etc.) Techniques not mutually exclusive Structural techniques stress testing - test larger-than-normal capacity in terms of transactions, data, users, speed, etc. execution testing- test performance in terms of speed, precision, etc. recovery testing - test how the system recovers from a disaster, how it handles corrupted data, etc. operations testing - test how the system fits in with existing operations and procedures in the user organization compliance testing - test adherence to standards security testing - test security requirements Functional techniques requirements testing - fundamental form of testing - makes sure the system does what its required to do regression testing - make sure unchanged functionality remains unchanged error-handling testing - test required error-handling functions (usually user error) manual-support testing - test that the system can be used properly - includes user documentation intersystem handling testing - test that the system is compatible with other systems in the environment control testing - test required control mechanisms parallel testing - feed same input into two versions of the system to make sure they produce the same output Unit Testing Goal is to evaluate some piece (file, program, module, component, etc.) in isolation Techniques can be structural or functional
- 46 -

In practice, its usually ad-hoc and looks a lot like debugging More structured approaches exist

8.4 Functional techniques

input domain testing - pick test cases representative of the range of allowable input, including high, low, and average values equivalence partitioning - partition the range of allowable input so that the program is expected to behave similarly for all inputs in a given partition, then pick a test case from each partition boundary value - choose test cases with input values at the boundary (both inside and outside) of the allowable range syntax checking - choose test cases that violate the format rules for input special values - design test cases that use input values that represent special situations output domain testing - pick test cases that will produce output at the extremes of the output domain Structural techniques statement testing - ensure the set of test cases exercises every statement at least once branch testing - each branch of an if/then statement is exercised conditional testing - each truth statement is exercised both true and false expression testing - every part of every expression is exercised path testing - every path is exercised (impossible in practice) Error-based techniques basic idea is that if you know something about the nature of the defects in the code, you can estimate whether or not youve found all of them or not fault seeding - put a certain number of known faults into the code, then test until they are all found mutation testing - create mutants of the program by making single changes, then run test cases until all mutants have been killed historical test data - an organization keeps records of the average numbers of defects in the products it produces, then tests a new product until the number of defects found approaches the expected number

8.5 Conclusion:
Hence the system Test phase should begin once modules are integrated enough to perform tests in a whole system environment. System testing can occur in parallel with integration test, especially with the top-down method.

- 47 -

Unit Testing

9.1 Introduction to Unit Testing

Unit testing. Isn't that some annoying requirement that we're going to ignore? Many developers get very nervous when you mention unit tests.Usually this is a vision of a grand table with every single method listed, along with the expected results and pass/fail date. It's important, but not relevant in most programming projects. The unit test will motivate the code that you write. In a sense, it is a little design document that says, "What will this bit of code do?" Or, in the language of object oriented programming, "What will these clusters of objects do?" The crucial issue in constructing a unit test is scope. If the scope is too narrow, then the tests will be trivial and the objects might pass the tests, but there will be no design of their interactions. Certainly, interactions of objects are the crux of any object oriented design. Likewise, if the scope is too broad, then there is a high chance that not every component of the new code will get tested. The programmer is then reduced to testing-by-poking-around, which is not an effective test strategy.

Need for Unit Test

How do you know that a method doesn't need a unit test? First, can it be tested by inspection? If the code is simple enough that the developer can just look at it and verify its correctness then it is simple enough to not require a unit test. The developer should know when this is the case. Unit tests will most likely be defined at the method level, so the art is to define the unit test on the methods that cannot be checked by inspection. Usually this is the case when the method involves a cluster of objects. Unit tests that isolate clusters of objects for testing are doubly useful, because they test for failures, and they also identify those segments of code that are related. People who revisit the code will use the unit tests to discover which objects are related, or which objects form a cluster. Hence: Unit tests isolate clusters of objects for future developers. Another good litmus test is to look at the code and see if it throws an error or catches an error. If error handling is performed in a method, then that method can break. Generally, any method that can break is a good candidate for having a unit test, because it may break at some time, and then the unit test will be there to help you fix it. The danger of not implementing a unit test on every method is that the coverage may be incomplete. Just because we don't test every method explicitly doesn't mean that methods can get away with not being tested. The programmer should know that their unit testing is complete when the unit tests cover at the very least - 48 -

the functional requirements of all the code. The careful programmer will know that their unit testing is complete when they have verified that their unit tests cover every cluster of objects that form their application.

Life Cycle Approach to Testing

Testing will occur throughout the project lifecycle i.e., from Requirements till User Acceptance Testing.The main Objective to Unit Testing are as follows : To execute a program with the intent of finding an error.; To uncover an as-yet undiscovered error ; and Prepare a test case with a high probability of finding an as-yet undiscovered error.

Levels of Unit Testing


Concepts in Unit Testing: The most 'micro' scale of testing; To test particular functions or code modules. Typically done by the programmer and not by testers. As it requires detailed knowledge of the internal program design and code. Not always easily done unless the application has a well-designed architecture with tight code;

9.2 Unit Testing Flow: driver

Module interface local data structures boundary conditions independent paths error handling paths TestCases

- 49 -

Types of Errors detected

The following are the Types of errors that may be caught Error in Data Structures Performance Errors Logic Errors Validity of alternate and exception flows Identified at analysis/design stages

Unit Testing Black Box Approach Field Level Check

Field Level Validation User Interface Check Functional Level Check

Unit Testing White Box Approach



Null / Not Null Checks Uniqueness Checks Length Checks Date Field Checks Numeric Checks Negative Checks

Unit Testing Field Level Validations

Test all Validations for an Input field Date Range Checks (From Date/To Dates) Date Check Validation with System date

Unit Testing User Interface Checks Readability of the Controls Tool Tips Validation Ease of Use of Interface Across Tab related Checks User Interface Dialog GUI compliance checks
- 50 -

Unit Testing - Functionality Checks

Screen Functionalities Field Dependencies Auto Generation Algorithms and Computations Normal and Abnormal terminations Specific Business Rules if any..



9.3 Execution of Unit Tests

Design a test case for every statement to be executed. Select the unique set of test cases. This measure reports whether each executable statement is encountered. Also known as: line coverage, segment coverage and basic block coverage. Basic block coverage is the same as statement coverage except the unit of code measured is each sequence of non-branching statements.

Example of Unit Testing:

int invoice (int x, int y) { int d1, d2, s; if (x<=30) d2=100; else d2=90; s=5*x + 10 *y; if (s<200) d1=100; else if (s<1000) d1 = 95; else d1 = 80; return (s*d1*d2/10000); }

Unit Testing Flow :

- 51 -

- 52 -

Advantage of Unit Testing Can be applied directly to object code and does not require processing source code. Performance profilers commonly implement this measure.

DISADVANTAGE of Unit Testing

Insensitive to some control structures (number of iterations)

Does not report whether loops reach their termination condition Statement coverage is completely insensitive to the logical operators (|| and &&).

Method for Statement Coverage -Design a test-case for the pass/failure of every decision point -Select unique set of test cases This measure reports whether Boolean expressions tested in control structures (such as the if-statement and while-statement) evaluated to both true and false. The entire Boolean expression is considered one true-or-false predicate regardless of whether it contains logical-and or logical-or operators. Additionally, this measure includes coverage of switch-statement cases, exception handlers, and interrupt handlers Also known as: branch coverage, all-edges coverage, basis path coverage, decisiondecision-path testing "Basis path" testing selects paths that achieve decision coverage. ADVANTAGE: Simplicity without the problems of statement coverage DISADVANTAGE This measure ignores branches within boolean expressions which occur due to shortcircuit operators. Method for Condition Coverage: -Test if every condition (sub-expression) in decision for true/false -Select unique set of test cases. Reports the true or false outcome of each Boolean sub-expression, separated by logical-and and logical-or if they occur. - 53 -

Condition coverage measures the sub-expressions independently of each other. Reports whether every possible combination of boolean sub-expressions occurs. As with condition coverage, the sub-expressions are separated by logical-and and logical-or, when present. The test cases required for full multiple condition coverage of a condition are given by the logical operator truth table for the condition. DISADVANTAGE: Tedious to determine the minimum set of test cases required, especially for very complex Boolean expressions Number of test cases required could vary substantially among conditions that have similar complexity Condition/Decision Coverage is a hybrid measure composed by the union of condition coverage and decision coverage. It has the advantage of simplicity but without the shortcomings of its component measures This measure reports whether each of the possible paths in each function have been followed. A path is a unique sequence of branches from the function entry to the exit. Also known as predicate coverage. Predicate coverage views paths as possible combinations of logical conditions Path coverage has the advantage of requiring very thorough testing FUNCTION COVERAGE:

This measure reports whether you invoked each function or procedure. It is useful during preliminary testing to assure at least some coverage in all areas of the software. Broad, shallow testing finds gross deficiencies in a test suite quickly.
LOOP COVERAGE This measure reports whether you executed each loop body zero times, exactly once, twice and more than twice (consecutively). For do-while loops, loop coverage reports whether you executed the body exactly once, and more than once. The valuable aspect of this measure is determining whether while-loops and for-loops execute more than once, information not reported by others measure. RACE COVERAGE This measure reports whether multiple threads execute the same code at the same time. Helps detect failure to synchronize access to resources. Useful for testing multi-threaded programs such as in an operating system.

9.4 Conclusion
Testing irrespective of the phases of testing should encompass the following : Cost of Failure associated with defective products getting shipped and used by customer is enormous
- 54 -

To find out whether the integrated product work as per the customer requirements To evaluate the product with an independent perspective To identify as many defects as possible before the customer finds To reduce the risk of releasing the product

- 55 -

10 Test Strategy
10.1 Introduction
This Document entails you towards the better insight of the Test Strategy and its methodology. It is the role of test management to ensure that new or modified service products meet the business requirements for which they have been developed or enhanced. The Testing strategy should define the objectives of all test stages and the techniques that apply. The testing strategy also forms the basis for the creation of a standardized documentation set, and facilitates communication of the test process and its implications outside of the test discipline. Any test support tools introduced should be aligned with, and in support of, the test strategy. Test Approach/Test Architecture are the acronyms for Test Strategy. Test management is also concerned with both test resource and test environment management.

10.2 Key elements of Test Management:

Test organization the set-up and management of a suitable test organizational structure and explicit role definition. The project framework under which the testing activities will be carried out is reviewed, high level test phase plans prepared and resource schedules considered. Test organization also involves the determination of configuration standards and the definition of the test environment. Test planning the requirements definition and design specifications facilitate in the identification of major test items and these may necessitate the test strategy to be updated. A detailed test plan and schedule is prepared with key test responsibilities being indicated. Test specifications required for all levels of testing and covering all categories of test. The required outcome of each test must be known before the test is attempted. Unit, integration and system testing configuration items are verified against the appropriate specifications and in accordance with the test plan. The test environment should also be under configuration control and test data and results stored for future evaluation. Test monitoring and assessment ongoing monitoring and assessment of the integrity of the development and construction. The status of the configuration items should be reviewed against the phase plans and test progress reports prepared providing some assurance of the verification and validation activities. Product assurance the decision to negotiate the acceptance testing program and the release and commissioning of the service product is subject to the product assurance role being satisfied with the outcome of the verification activities. Product assurance may oversee some of the test activity and may participate in process reviews. A common criticism of construction programmers is that insufficient time is frequently allocated to the testing and commissioning of the building systems together with the involvement and subsequent training of the Facilities Management team. Testing and commissioning is often considered by teams as a secondary activity and given a lower priority particularly as pressure builds on the program towards completion. Sufficient time must be dedicated to testing and commissioning as ensuring the systems function correctly is fairly fundamental to the projects success or failure. Traditionally the responsibility for testing and commissioning is buried deep within the supply chain as a sub-contract of a sub-contract. - is possible to gain greater control of this process and It - 56

the associated risk through the use of specialists such as Systems Integration who can be appointed as part of the professional team. The time necessary for testing and commissioning will vary from project to project depending upon the complexity of the systems and services that have been installed. The Project Sponsor should ensure that the professional team and the contractor consider realistically how much time is needed. Fitness for purpose checklist: Is there a documented testing strategy that defines the objectives of all test stages and the techniques that may apply, e.g. non-functional testing and the associated techniques such as performance, stress and security etc? Does the test plan prescribe the approach to be taken for intended test activities, identifying: the items to be tested, the testing to be performed, test schedules, resource and facility requirements, reporting requirements, evaluation criteria, risks requiring contingency measures? Are test processes and practices reviewed regularly to assure that the testing processes continue to meet specific business needs? For example, e-commerce testing may involve new user interfaces and a business focus on usability may mean that the organization must review its testing strategies.

10.3 Test Strategy Flow :

Test Cases and Test Procedures should manifest Test Strategy.

- 57 -

Test Strategy Selection

Selection of the Test Strategy is based on the following factors Product Test Strategy based on the Application to help people and teams of people in making decisions. Based on the Key Potential Risks Suggestion of Wrong Ideas. People will use the Product Incorrectly Incorrect comparison of scenarios. Scenarios may be corrupted. Unable to handle Complex Decisions. Determination of Actual Risk. Understand the underlying Algorithm. Simulate the Algorithm in parallel. Capability test each major function. Generate large number of decision scenarios. Create complex scenarios and compare them. Review Documentation and Help. Test for sensitivity to user Error.

Test Strategy Execution:

Understand the decision Algorithm and generate the parallel decision analyzer using the Perl or Excel that will function as a reference for high volume testing of the app. Create a means to generate and apply large numbers of decision scenarios to the product. This will be done using the GUI test Automation system or through the direct generation of Decide Right scenario files that would be loaded into the product during test. Review the Documentation, and the design of the user interface and functionality for its sensitivity to user error. Test with decision scenarios that are near the limit of complexity allowed by the product Compare complex scenarios. Test the product for the risk of silent failures or corruptions in decision analysis. Issues in Execution of the Test Strategy The difficulty of understanding and simulating the decision algorithm The risk of coincidal failure of both the simulation and the product. The difficulty of automating decision tests

10.4 General Testing Strategies

Top-down Bottom-up Thread testing Stress testing Back-to-back testing

- 58 -

10.5 Need for Test Strategy

The objective of testing is to reduce the risks inherent in computer systems. The strategy must address the risks and present a process that can reduce those risks. The system concerns on risks then establish the objectives for the test process. The two components of the testing strategy are the Test Factors and the Test Phase.

Analysis Errors 36% and design Errors 64%


Test Factor The risk or issue that needs to be addressed as part of the test strategy. The strategy will select those factors that need to be addressed in the testing of a specific application system. Test Phase The Phase of the systems development life cycle in which testing will occur. Not all the test factors will be applicable to all software systems. The development team will need to select and rank the test factors for the specific software systems being developed. The test phase will vary based on the testing methodology used. For example the test phases in as traditional waterfall life cycle methodology will be much different from the phases in a Rapid Application Development methodology.

10.6 Developing a Test Strategy

The test Strategy will need to be customized for any specific software system. The applicable test factors would be listed as the phases in which the testing must occur. Four test steps must be followed to develop a customized test strategy. Select and rank Test Factors Identify the System Developmental Phases Identify the Business risks associated with the System under Development. Place risks in the Matrix
TestFac tors\Te st Phase Requir ements Design Build Dynami c Test Integra te Maintai n

Facto rs
- 59 -


10.7 Conclusion:
Test Strategy should be developed in accordance with the business risks associated with the software when the test team develop the test tactics. Thus the Test team needs to acquire and study the test strategy that should question the following: What is the relationship of importance among the test factors? Which of the high level risks are the most significant? What damage can be done to the business if the software fails to perform correctly? What damage can be done to the business if the business if the software is not completed on time? Who are the individuals most knowledgeable in understanding the impact of the identified business risks?

Hence the Test Strategy must address the risks and present a process that can reduce those risks. The system accordingly focuses on risks thereby establishes the objectives for the test process.

- 60 -

11.1 What is a Test Plan?
A Test Plan can be defined as a document that describes the scope, approach, resources and schedule of intended test activities. It identifies test items, the features to be tested, the testing tasks, who will do each task, and any risks requiring contingency planning. The main purpose of preparing a Test Plan is that everyone concerned with the project are in sync with regards to the scope, responsibilities, deadlines and deliverables for the project. It is in this respect that reviews and a sign-off are very important since it means that everyone is in agreement of the contents of the test plan and this also helps in case of any dispute during the course of the project (especially between the developers and the testers).

Purpose of preparing a Test Plan

A Test Plan is a useful way to think through the efforts needed to validate the acceptability of a software product. The completed document will help people outside the test group understand the 'why' and 'how' of product validation. It should be thorough enough to be useful but not so thorough that no one outside the test group will read it.

Contents of a Test Plan

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. Purpose Scope Test Approach Entry Criteria Resources Tasks / Responsibilities Exit Criteria Schedules / Milestones Hardware / Software Requirements Risks & Mitigation Plans Tools to be used Deliverables References a. Procedures b. Templates c. Standards/Guidelines 14. Annexure 15. Sign-Off

11.2 Contents (in detail)

Purpose This section should contain the purpose of preparing the test plan

- 61 -

This section should talk about the areas of the application which are to be tested by the QA team and specify those areas which are definitely out of scope (screens, database, mainframe processes etc). Test Approach This would contain details on how the testing is to be performed and whether any specific strategy is to be followed (including configuration management). Entry Criteria This section explains the various steps to be performed before the start of a test (i.e.) pre-requisites. For example: Timely environment set up, starting the web server / app server, successful implementation of the latest build etc. Resources This section should list out the people who would be involved in the project and their designation etc. Tasks / Responsibilities This section talks about the tasks to be performed and the responsibilities assigned to the various members in the project. Exit criteria Contains tasks like bringing down the system / server, restoring system to pre-test environment, database refresh etc. Schedules / Milestones This sections deals with the final delivery date and the various milestone dates to be met in the course of the project. Hardware / Software Requirements This section would contain the details of PCs / servers required (with the configuration) to install the application or perform the testing; specific software that needs to be installed on the systems to get the application running or to connect to the database; connectivity related issues etc. Risks & Mitigation Plans This section should list out all the possible risks that can arise during the testing and the mitigation plans that the QA team plans to implement incase the risk actually turns into a reality. Tools to be used This would list out the testing tools or utilities (if any) that are to be used in the project (e.g.) WinRunner, Test Director, PCOM, WinSQL. Deliverables This section contains the various deliverables that are due to the client at various points of time (i.e.) daily, weekly, start of the project, end of the project etc. These could include Test Plans, Test Procedure, Test Matrices, Status Reports, Test Scripts etc. Templates for all these could also be attached. References Procedures

- 62 -

Templates (Client Specific or otherwise) Standards / Guidelines (e.g.) QView Project related documents (RSD, ADD, FSD etc) Annexure This could contain embedded documents or links to documents which have been / will be used in the course of testing (e.g.) templates used for reports, test cases etc. Referenced documents can also be attached here. Sign-Off This should contain the mutual agreement between the client and the QA team with both leads / managers signing off their agreement on the Test Plan.

- 63 -

12 Test Data Preparation - Introduction

A System is programmed by its data. Functional testing can suffer if data is poor, and good data can help improve functional testing. Good test data can be structured to improve understanding and testability. Its contents, correctly chosen, can reduce maintenance effort and allow flexibility. Preparation of the data can help to focus the business where requirements are vague. The first stage of any recogniser development project is data preparation. Test data should however, be prepared which is representative of normal business transactions. Actual customer names or contact details should also not be used for such tests. It is recommended that a full test environment be set up for use in the applicable circumstances. Each separate test should be given a unique reference number which will identify the Business Process being recorded, the simulated conditions used, the persons involved in the testing process and the date the test was carried out. This will enable the monitoring and testing reports to be co-coordinated with any feedback received. Tests must be planned and thought out a head of time; you have to decide such things as what exactly you are testing and testing for, the way the test is going to be run and applied, what steps are required, etc. Testing is the process of creating, implementing and evaluating tests. Effective quality control testing requires some basic goals and understanding: You must understand what you are testing; if you're testing a specific functionality, you must know how it's supposed to work, how the protocols behave, etc. You should have a definition of what success and failure are. In other words, is close enough good enough? You should have a good idea of a methodology for the test, the more formal a plan the better; you should design test cases. You must understand the limits inherent in the tests themselves. You must have a consistent schedule for testing; performing a specific set of tests at appropriate points in the process is more important than running the tests at a specific time. Roles of Data in Functional Testing Testing consumes and produces large amounts of data. Data describes the initial conditions for a test, forms the input, is the medium through which the tester influences the software. Data is manipulated, extrapolated, summarized and referenced by the functionality under test, which finally spews forth yet more data to be checked against expectations. Data is a crucial part of most functional testing. This paper sets out to illustrate some of the ways that data can influence the test process, and will show that testing can be improved by a careful choice of input data. In doing this, the paper will concentrate most on data-heavy applications; those which use databases or are heavily influenced by the data they hold. The paper will focus on input data, rather than output data or the transitional states the data passes through during processing, as input data has the greatest influence on functional testing and is the simplest to manipulate. The paper will not consider areas where data is important to nonfunctional testing, such as operational profiles, massive datasets and environmental tuning. A SYSTEM IS PROGRAMMED BY ITS DATA Many modern systems allow tremendous flexibility in the way their basic functionality can be used. Configuration data can dictate control flow, data manipulation, presentation and user interface. A system can be configured to fit several business models, work (almost)
- 64 -

seamlessly with a variety of cooperative systems and provide tailored experiences to a host of different users. A business may look to an application's configurability to allow them to keep up with the market without being slowed by the development process, an individual may look for a personalized experience from commonly-available software. FUNCTIONAL TESTING SUFFERS IF DATA IS POOR Tests with poor data may not describe the business model effectively, they may be hard to maintain, or require lengthy and difficult setup. They may obscure problems or avoid them altogether. Poor data tends to result in poor tests, that take longer to execute. GOOD DATA IS VITAL TO RELIABLE TEST RESULTS An important goal of functional testing is to allow the test to be repeated with the same result, and varied to allow diagnosis. Without this, it is hard to communicate problems to coders, and it can become difficult to have confidence in the QA team's results, whether they are good or bad. Good data allows diagnosis, effective reporting, and allows tests to be repeated with confidence,. GOOD DATA CAN HELP TESTING STAY ON SCHEDULE An easily comprehensible and well-understood dataset is a tool to help communication. Good data can greatly assist in speedy diagnosis and rapid re-testing. Regression testing and automated test maintenance can be made speedier and easier by using good data, while an elegantly-chosen dataset can often allow new tests without the overhead of new data. A formal test plan is a document that provides and records important information about a test project, for example: project and quality assumptions project background information resources schedule & timeline entry and exit criteria test milestones tests to be performed

use cases and/or test cases

12.1 Criteria for Test Data Collection

This section of the Document specifies the description of the test data needed to test recovery of each business process. Identify Who is to Conduct the Tests In order to ensure consistency of the testing process throughout the organization, one or more members of the Business Continuity Planning (BCP) Team should be nominated to co-ordinate the testing process within each business unit, a nominated testing and across the organization. Each business process should be thoroughly tested and the coordinator should ensure that each business unit observes the necessary rules associated with ensuring that the testing process is carried out within a realistic environment. This section of the BCP should contain the names of the BCP Team members nominated to co-ordinate the testing process. It should also list the duties of the appointed coordinators.

Identify Who is to Control and Monitor the Tests

In order to ensure consistency when measuring the results, the tests should be independently monitored. This task would normally be carried out by a nominated member of the Business Recovery Team or a member of the Business Continuity Planning Team. - 65 -

This section of the BCP will contain the names of the persons nominated to monitor the testing process throughout the organization. It will also contain a list of the duties to be undertaken by the monitoring staff. Prepare Feedback Questionnaires It is vital to receive feedback from the persons managing and participating in each of the tests. This feedback will hopefully enable weaknesses within the Business Recovery Process to be identified and eliminated. Completion of feedback forms should be mandatory for all persons participating in the testing process. The forms should be completed either during the tests (to record a specific issue) or as soon after finishing as practical. This will enable observations and comments to be recorded whilst the event is still fresh in the persons mind. This section of the BCP should contain a template for a Feedback Questionnaire. Prepare Budget for Testing Phase Each phase of the BCP process which incurs a cost requires that a budget be prepared and approved. The 'Preparing for a Possible Emergency' Phase of the BCP process will involve the identification and implementation of strategies for back up and recovery of data files or a part of a business process. It is inevitable that these back up and recovery processes will involve additional costs. Critical parts of the business process such as the IT systems, may require particularly expensive back up strategies to be implemented. Where the costs are significant they should be approved separately with a specific detailed budget for the establishment costs and the ongoing maintenance costs. This section of the BCP will contain a list of the testing phase activities and a cost for each. It should be noted whenever part of the costs is already incorporated with the organizations overall budgeting process.

Training Core Testing Team for each Business Unit

In order for the testing process to proceed smoothly, it is necessary for the core testing team to be trained in the emergency procedures. This is probably best handled in a workshop environment and should be presented by the persons responsible for developing the emergency procedures. This section of the BCP should contain a list of the core testing team for each of the business units who will be responsible for coordinating and undertaking the Business Recovery Testing process. It is important that clear instructions are given to the Core Testing Team regarding the simulated conditions which have to be observed. Conducting the Tests The tests must be carried out under authentic conditions and all participants must take the process seriously. It is important that all persons who are likely to be involved with recovering a particular business process in the event of an emergency should participate in the testing process. It should be mandatory for the management of a business unit to be present when that unit is involved with conducting the tests. Test each part of the Business Recovery Process In so far as it is practical, each critical part of the business recovery process should be fully tested. Every part of the procedures included as part of the recovery process is to be tested to ensure validity and relevance. This section of the BCP is to contain a list of each business process with a test schedule and information on the simulated conditions being used. The testing co-ordination and monitoring will endeavor to ensure that the simulated environments are maintained throughout the testing process, in a realistic manner. Test Accuracy of Employee -and Vendor Emergency Contact Numbers 66 -

During the testing process the accuracy of employee and vendor emergency contact information is to be re-confirmed. All contact numbers are to be validated for all involved employees. This is particularly important for management and key employees who are critical to the success of the recovery process. This activity will usually be handled by the HRM Department or Division. Where, in the event of an emergency occurring outside of normal business hours, a large number of persons are to be contacted, a hierarchical process could be used whereby one person contacts five others. This process must have safety features incorporated to ensure that if one person is not contactable for any reason then this is notified to a nominated controller. This will enable alternative contact routes to be used. Assess Test Results Prepare a full assessment of the test results for each business process. The following questions may be appropriate: Were objectives of the Business Recovery Process and the testing process met - if not, provide further comment Were simulated conditions reasonably "authentic" - if not, provide further comment Was test data representative - if not, provide further comment Did the tests proceed without any problems - if not, provide further comment What were the main comments received in the feedback questionnaires Each test should be assessed as either fully satisfactory, adequate or requiring further testing. Training Staff in the Business Recovery Process All staff should be trained in the business recovery process. This is particularly important when the procedures are significantly different from those pertaining to normal operations. This training may be integrated with the training phase or handled separately. The training should be carefully planned and delivered on a structured basis. The training should be assessed to verify that it has achieved its objectives and is relevant for the procedures involved. Training may be delivered either using in-house resources or external resources depending upon available skills and related costs. Managing the Training Process For the BCP training phase to be successful it has to be both well managed and structured. It will be necessary to identify the objective and scope for the training, what specific training is required, who needs it and a budget prepared for the additional costs associated with this phase. Develop Objectives and Scope of Training The objectives and scope of the BCP training activities are to be clearly stated within the plan. The BCP should contain a description of the objectives and scope of the training phase. This will enable the training to be consistent and organized in a manner where the results can be measured, and the training fine tuned, as appropriate. The objectives for the training could be as follows : "To train all staff in the particular procedures to be followed during the business recovery process". The scope of the training could be along the following lines : "The training is to be carried out in a comprehensive and exhaustive manner so that staff become familiar with all aspects of the recovery process. The training will cover all aspects of the Business Recovery activities section of the BCP including IT systems recovery". - 67 -

Consideration should also be given to the development of a comprehensive corporate awareness program for communicating the procedures for the business recovery process. Training Needs Assessment The plan must specify which person or group of persons requires which type of training. It is necessary for all new or revised processes to be explained carefully to the staff. For example it may be necessary to carry out some process manually if the IT system is down for any length of time. These manual procedures must be fully understood by the persons who are required to carry them out. For larger organizations it may be practical to carry out the training in a classroom environment, however, for smaller organizations the training may be better handled in a workshop style. This section of the BCP will identify for each business process what type of training is required and which persons or group of persons need to be trained. Training Materials Development Schedule Once the training needs have been identified it is necessary to specify and develop suitable training materials. This can be a time consuming task and unless priorities are given to critical training programmes, it could delay the organization in reaching an adequate level of preparedness. This section of the BCP contains information on each of the training programmes with details of the training materials to be developed, an estimate of resources and an estimate of the completion date. Prepare Training Schedule Once it has been agreed who requires training and the training materials have been prepared a detailed training schedule should be drawn up. This section of the BCP contains the overview of the training schedule and the groups of persons receiving the training. Communication to Staff Once the training is arranged to be delivered to the employees, it is necessary to advise them about the training programmes they are scheduled to attend. This section of the BCP contains a draft communication to be sent to each member of staff to advise them about their training schedule. The communication should provide for feedback from the staff member where the training dates given are inconvenient. A separate communication should be sent to the managers of the business units advising them of the proposed training schedule to be attended by their staff. Each member of staff will be given information on their role and responsibilities applicable in the event of an emergency. Prepare Budget for Training Phase Each phase of the BCP process which incurs a cost requires that a budget be prepared and approved. Depending upon the cross charging system employed by the organization, the training costs will vary greatly. However, it has to be recognized that, however well justified, training incurs additional costs and these should be approved by the appropriate authority within the organization. This section of the BCP will contain a list of the training phase activities and a cost for each. It should be noted whenever part of the costs is already incorporated with the organizations overall budgeting process. Assessing the Training The individual BCP training programmes and the overall BCP training process should be assessed to ensure its effectiveness and applicability. This information will be gathered from the trainers and also the trainees through the completion of feedback questionnaires. Feedback Questionnaires Assess Feedback - 68 -

Feedback Questionnaires It is vital to receive feedback from the persons managing and participating in each of the training programmes. This feedback will enable weaknesses within the Business Recovery Process, or the training, to be identified and eliminated. Completion of feedback forms should be mandatory for all persons participating in the training process. The forms should be completed either during the training (to record a specific issue) or as soon after finishing as practical. This will enable observations and comments to be recorded whilst the event is still fresh in the persons mind. This section of the BCP should contain a template for a Feedback Questionnaire for the training phase. Assess Feedback The completed questionnaires from the trainees plus the feedback from the trainers should be assessed. Identified weaknesses should be notified to the BCP Team Leader and the process strengthened accordingly. The key issues raised by the trainees should be noted and consideration given to whether the findings are critical to the process or not. If there are a significant number of negative issues raised then consideration should be given to possible re-training once the training materials, or the process, have been improved. This section of the BCP will contain a format for assessing the training feedback. Keeping the Plan Up-to-date Changes to most organizations occur all the time. Products and services change and also their method of delivery. The increase in technological based processes over the past ten years, and particularly within the last five, have significantly increased the level of dependency upon the availability of systems and information for the business to function effectively. These changes are likely to continue and probably the only certainty is that the pace of change will continue to increase. It is necessary for the BCP to keep pace with these changes in order for it to be of use in the event of a disruptive emergency. This chapter deals with updating the plan and the managed process which should be applied to this updating activity. Maintaining the BCP It is necessary for the BCP updating process to be properly structured and controlled. Whenever changes are made to the BCP they are to be fully tested and appropriate amendments should be made to the training materials. This will involve the use of formalized change control procedures under the control of the BCP Team Leader. Change Controls for Updating the Plan It is recommended that formal change controls are implemented to cover any changes required to the BCP. This is necessary due to the level of complexity contained within the BCP. A Change request Form / Change Order form is to be prepared and approved in respect of each proposed change to the BCP. This section of the BCP will contain a Change Request Form / Change Order to be used for all such changes to the BCP. Responsibilities for Maintenance of Each Part of the Plan Each part of the plan will be allocated to a member of the BCP Team or a Senior Manager with the organization who will be charged with responsibility for updating and maintaining the plan. The BCP Team Leader will remain in overall control of the BCP but business unit heads will need to keep their own sections of the BCP up to date at all times. Similarly, HRM Department will be responsible to ensure that all emergency contact numbers for staff are kept up to date. It is important that the relevant BCP coordinator and the Business Recovery Team are kept fully informed regarding any approved changes to the plan. Test All Changes to Plan
- 69 -

The BCP Team will nominate one or more persons who will be responsible for coordinating all the testing processes and for ensuring that all changes to the plan are properly tested. Whenever changes are made or proposed to the BCP, the BCP Testing Co-ordinator will be notified. The BCP Testing Co-ordinator will then be responsible for notifying all affected units and for arranging for any further testing activities. This section of the BCP contains a draft communication from the BCP Co-ordinator to affected business units and contains information about the changes which require testing or re-testing. Advise Person Responsible for BCP Training A member of the BCP Team will be given responsibility for co-ordinating all training activities (BCP Training Co-ordinator). The BCP Team Leader will notify the BCP Training Co-ordinator of all approved changes to the BCP in order that the training materials can be updated. An assessment should be made on whether the change necessitates any re-training activities. Advise Person Responsible for BCP Training A member of the BCP Team will be given responsibility for co-ordinating all training activities (BCP Training Co-ordinator). The BCP Team Leader will notify the BCP Training Co-ordinator of all approved changes to the BCP in order that the training materials can be updated. An assessment should be made on whether the change necessitates any re-training activities. Problems which can be caused by Poor Test Data Most testers are familiar with the problems that can be caused by poor data. The following list details the most common problems familiar to the author. Most projects experience these problems at some stage - recognizing them early can allow their effects to be mitigated. Unreliable test results. Running the same test twice produces inconsistent results. This can be a symptom of an uncontrolled environment, unrecognized database corruption, or of a failure to recognize all the data that is influential on the system. Degradation of test data over time. Program faults can introduce inconsistency or corruption into a database. If not spotted at the time of generation, they can cause hard-to-diagnose failures that may be apparently unrelated to the original fault. Restoring the data to a clean set gets rid of the symptom, but the original fault is undiagnosed and can carry on into live operation and perhaps future releases. Furthermore, as the data is restored, evidence of the fault is lost. Increased test maintenance cost If each test has its own data, the cost of test maintenance is correspondingly increased. If that data is itself hard to understand or manipulate, the cost increases further. Reduced flexibility in test execution If datasets are large or hard to set up, some tests may be excluded from a test run. If the datasets are poorly constructed, it may not be time-effective to construct further data to support investigatory tests. Obscure results and bug reports Without clearly comprehensible data, testers stand a greater chance of missing important diagnostic features of a failure, or indeed of missing the failure entirely. Most reports make reference to the input data and the actual and expected results. Poor data can make these reports hard to understand. Larger proportion of problems can be traced to poor data
- 70 -

A proportion of all failures logged will be found, after further analysis, not to be faults at all. Data can play a significant role in these failures. Poor data will cause more of these problems. Less time spent hunting bugs The more time spent doing unproductive testing or ineffective test maintenance, the less time spent testing. Confusion between developers, testers and business Each of these groups has different data requirements. A failure to understand each others data can lead to ongoing confusion. Requirements problems can be hidden in inadequate data It is important to consider inputs and outputs of a process for requirements modeling. Inadequate data can lead to ambiguous or incomplete requirements. Simpler to make test mistakes Everybody makes mistakes. Confusing or over-large datasets can make data selection mistakes more common. Unwieldy volumes of data Small datasets can be manipulated more easily than large datasets. A few datasets are easier to manage than many datasets. Business data not representatively tested Test requirements, particularly in configuration data, often don't reflect the way the system will be used in practice. While this may arguably lead to broad testing for a variety of purposes, it can be hard for the business or the end users to feel confidence in the test effort if they feel distanced from it. Inability to spot data corruption caused by bugs A few well-known datasets can be more easily be checked than a large number of complex datasets, and may lend themselves to automated testing / sanity checks. A readily understandable dataset can allow straightforward diagnosis; a complex dataset will positively hinder diagnosis. Poor database/environment integrity If a large number of testers, or tests, share the same dataset, they can influence and corrupt each others results as they change the data in the system. This can not only cause false results, but can lead to database integrity problems and data corruption. This can make portions of the application untestable for many testers simultaneously.

12.2 Classification of Test Data Types

In the process of testing a system, many references are made to "The Data" or "Data Problems". Although it is perhaps simpler to discuss data in these terms, it is useful to be able to classify the data according to the way it is used. The following broad categories allow data to be handled and discussed more easily. Environmental data Environmental data tells the system about its technical environment. It includes communications addresses, directory trees and paths and environmental variables. The current date and time can be seen as environmental data. Setup data
- 71 -

Setup data tells the system about the business rules. It might include a cross reference between country and delivery cost or method, or methods of debt collection from different kinds of customers. Typically, setup data causes different functionality to apply to otherwise similar data. With an effective approach to setup data, business can offer new intangible products without developing new functionality - as can be seen in the mobile phone industry, where new billing products are supported and indeed created by additions to the setup data. Input data Input data is the information input by day-to-day system functions. Accounts, products, orders, actions, documents can all be input data. For the purposes of testing, it is useful to split the categorization once more: FIXED INPUT DATA Fixed input data is available before the start of the test, and can be seen as part of the test conditions. CONSUMABLE INPUT DATA Consumable input data forms the test input It can also be helpful to qualify data after the system has started to use it; Transitional data Transitional data is data that exists only within the program, during processing of input data. Transitional data is not seen outside the system (arguably, test handles and instrumentation make it output data), but its state can be inferred from actions that the system has taken. Typically held in internal system variables, it is temporary and is lost at the end of processing. Output data Output data is all the data that a system outputs as a result of processing input data and events. It generally has a correspondence with the input data (cf. Jackson's Structured Programming methodology), and includes not only files, transmissions, reports and database updates, but can also include test measurements. A subset of the output data is generally compared with the expected results at the end of test execution. As such, it does not directly influence the quality of the tests.

12.3 Organizing the data

A key part of any approach to data is the way the data is organized; the way it is chosen and described, influenced by the uses that are planned for it. A good approach increases data reliability, reduces data maintenance time and can help improve the test process. Good data assists testing, rather than hinders it. Permutations Most testers are familiar with the concept of permutation; generating tests so that all possible permutations of inputs are tested. Most are also familiar with the ways in which this generally vast set can be cut down.
- 72 -

Pair wise, or combinatorial testing addresses this problem by generating a set of tests that allow all possible pairs of combinations to be tested. Typically, for non-trivial sets, this produces a far smaller set of tests than the brute-force approach for all permutations, The same techniques can be applied to test data; the test data can contain all possible pairs of permutations in a far smaller set than that which contains all possible permutations. This allows a small, easy to handle dataset - which also allows a wide range of tests. This small, and easy to manipulate dataset is capable of supporting many tests. It allows complete pairwise coverage, and so is comprehensive enough to allow a great many new, ad-hoc, or diagnostic tests. Database changes will affect it, but the data maintenance required will be greatly lessened by the small size of the dataset and the amount of reuse it allows. Finally, this method of working with fixed input data can help greatly in testing the setup data. This method is most appropriate when used, as above, on fixed input data. It is most effective when the following conditions are satisfied. Fortunately, these criteria apply to many traditional database-based systems: ixed input data consists of many rows F Fields are independent You want to do many tests without loading / you do not load fixed input data for each test. To sum up, permutation helps because: Permutation is familiar from test planning. Achieves good test coverage without having to construct massive datasets Can perform investigative testing without having to set up more data Reduces the impact of functional/database changes Can be used to test other data - particularly setup data

Partitions allow data access to be controlled, reducing uncontrolled changes in the data. Partitions can be used independently; data use in one area will have no effect on the results of tests in another. Data can be safely and effectively partitioned by machine / database / application instance, although this partitioning can introduce configuration management problems in software version, machine setup, environmental data and data load/reload. A useful and basic way to start with partitions is to set up, not a single environment for each test or tester, but to set up three shared by many users, so allowing different kinds of data use. These three have the following characteristics: Safe area Used for enquiry tests, usability tests etc. No test changes the data, so the area can be trusted. Many testers can use simultaneously Change area Used for tests which update/change data. Data must be reset or reloaded after testing. Used by one test/tester at a time. Scratch area Used for investigative update tests and those which have unusual requirements. Existing data cannot be trusted. Used at tester's own risk! Testing rarely has the luxury of completely separate environments for each test and each tester. - 73 -

Controlling data, and the access to data, in a system can be fraught. Many different stakeholders have different requirements of the data, but a common requirement is that of exclusive use. While the impact of this requirement should not be underestimated, a number of stakeholders may be able to work with the same environmental data, and to a lesser extent, setup data - and their work may not need to change the environmental or setup data. The test strategy can take advantage of this by disciplined use of text / value fields, allowing the use of 'soft' partitions. 'Soft' partitions allow the data to be split up conceptually, rather than physically. Although testers are able to interfere with each others tests, the team can be educated to avoid each others work. If, for instance, tester 1's tests may only use customers with Russian nationality and tester 2's tests only with French, the two sets of work can operate independently in the same dataset. A safe area could consist of London addresses, the change area Manchester addresses, and the scratch area Bristol addresses. Typically, values in free-text fields are used for soft partitioning. Data partitions help because: Allow controlled and reliable data, reducing data corruption / change problems Can reduce the need for exclusive access to environments/machines Clarity Permutation techniques may make data easier to grasp by making the datasets small and commonly used, but we can make our data clearer still by describing each row in its own free text fields, allowing testers to make a simple comparison between the free text (which is generally displayed on output), and actions based on fields which tend not to be directly displayed. Use of free text fields with some correspondence to the internals of the record allows output to be checked more easily. Testers often talk about items of data, referring to them by anthropomorphic personification - that is to say, they give them names. This allows shorthand, but also acts as jargon, excluding those who are not in the know. Setting this data, early on in testing, to have some meaningful value can be very useful, allowing testers to sense check input and output data, and choose appropriate input data for investigative tests. Reports, data extracts and sanity checks can also make use of these; sorting or selecting on a free text field that should have some correspondence with a functional field can help spot problems or eliminate unaffected data. Data is often used to communicate and illustrate problems to coders and to the business. However, there is generally no mandate for outside groups to understand the format or requirements of test data. Giving some meaning to the data that can be referred to directly can help with improving mutual understanding. Clarity helps because: Improves communication within and outside the team Reduces test errors caused by using the wrong data Allows another method way of doing sanity checks for corrupted or inconsistent data Helps when checking data after input Helps in selecting data for investigative tests

12.4 Data Load and Data Maintenance

An important consideration in preparing data for functional testing is the ways in which the data can be loaded into the system, and the possibility and ease of maintenance. Loading the data
- 74 -

Data can be loaded into a test system in three general ways. Using the system you're trying to test The data can be manually entered, or data entry can be automated by using a capture/replay tool. This method can be very slow for large datasets. It uses the system's own validation and insertion methods, and can both be hampered by faults in the system, and help pinpoint them. If the system is working well, data integrity can be ensured by using this method, and internally assigned keys are likely to be effective and consistent. Data can be well-described in test scripts, or constructed and held in flat files. It may, however, be input in an ad-hoc way, which is unlikely to gain the advantages of good data listed above. Using a data load tool Data load tools directly manipulate the system's underlying data structures. As they do not use the system's own validation, they can be the only way to get broken data into the system in a consistent fashion. As they do not use the system to load the data, they can provide a convenient workaround to known faults in the system's data load routines. However, they may come up against problems when generating internal keys, and can have problems with data integrity and parent/child relationships. Data loaded can have a range of origins. In some cases, all new data is created for testing. This data may be complete and well specified, but can be hard to generate. A common compromise is to use old data from an existing system, selected for testing, filtered for relevance and duplicates and migrated to the target data format. In some cases, particularly for minor system upgrades, the complete set of live data is loaded into the system, but stripped of personal details for privacy reasons. While this last method may seem complete, it has disadvantages in that the data may not fully support testing, and that the large volume of data may make test results hard to interpret. Not loaded at all Some tests simply take whatever is in the system and try to test with it. This can be appropriate where a dataset is known and consistent, or has been set up by a prior round of testing. It can also be appropriate in environments where data cannot be reloaded, such as the live system. However, it can be symptomatic of an uncontrolled approach to data, and is not often desirable. Environmental data tends to be manually loaded, either at installation or by manipulating environmental or configuration scripts. Large volumes of setup data can often be generated from existing datasets and loaded using a data load tool, while small volumes of setup data often have an associated system maintenance function and can be input using the system. Fixed input data may be generated or migrated and is loaded using any and all of the methods above, while consumable input data is typically listed in test scripts or generated as an input to automation tools. When data is loaded, it can append itself to existing data, overwrite existing data, or delete existing data first. Each is appropriate in different circumstances, and due consideration should be given to the consequences.

12.5 Testing the Data

A theme bought out at the start of this paper was 'A System is Programmed by its Data'. In order to test the system, one must also test the data it is configured with; the environmental and setup data. Environmental data is necessarily different between the test and live environment. Although testing can verify that the environmental variables are being read and used correctly, there is little point in testing their values on a system other than the target
- 75 -

system. Environmental data is often checked manually on the live system during implementation and rollout, and the wide variety of possible methods will not be discussed further here. Setup data can change often, throughout testing, as the business environment changes particularly if there is a long period between requirements gathering and live rollout. Testing done on the setup data needs to cover two questions; Does the planned/current setup data induce the functionality that the business requires? Will changes made to the setup data have the desired effect? Testing for these two questions only becomes possible when that data is controlled. Aspects of all the elements above come into play; The setup data should be organized to allow a good variety of scenarios to be considered The setup data needs to be able to be loaded and maintained easily and repeatable The business needs to become involved in the data so that their setup for live can be properly tested When testing the setup data, it is important to have a well-known set of fixed input data and consumable input data. This allows the effects of changes made to the setup data to be assessed repeat ably and allows results to be compared. The advantages of testing the setup data include: Overall testing will be improved if the quality of the setup data improves Problems due to faults in the live setup data will be reduced The business can re-configure the software for new business needs with increased confidence Data-related failures in the live system can be assessed in the light of good data testing

12.6 Conclusion
Data can be influential on the quality of testing. Well-planned data can allow flexibility and help reduce the cost of test maintenance. Common data problems can be avoided or reduced with preparation and automation. Effective testing of setup data is a necessary part of system testing, and good data can be used as a tool to enable and improve communication throughout the project. The following points summarize the actions that can influence the quality of the data and the effectiveness of its usage: Plan the data for maintenance and flexibility Know your data, and make its structure and content transparent Use the data to improve understanding throughout testing and the business Test setup data as you would test functionality

- 76 -

13 Test Logs - Introduction

Test Problem is a condition that exists within the software system that needs to be addressed. Carefully and completely documenting a test problem is the first step in correcting the problem. The following four attributes should be developed for all the test problems: Statement of condition. Tells what it is. Criteria Tells what should be. These two attributes are the basis for a finding. If a comparison between the two gives little or no practical consequence, no finding exists. Effect: Tells why the difference between what is and what should be is significant Cause: Tells the reasons for the deviation. Identification of the cause is the necessary as a basis for corrective action. A well developed problem statement will include each of these atttributes.When one or more these attributes is missing, questions almost arise, such as Criteria: Why is the current state inadequate? Effect: How significant is it? Cause: What could have cause of the problem?

13.1 Factors defining the Test Log Generation

Document Deviation: Problem statements begin to emerge by process of comparision.Essentially the user compares what is with what should be. When a deviation is identified between what is found to actually exist and what the user thinks is correct or proper , the first essential step toward development of a problem statement has occurred. It is difficult to visualize any type of problem that is not in some way characterized by this deviation. The What is: can be called the statement of condition. The What should be shall be called the Criteria. These concepts are the first two and the most basic , attributes of a problem statement. The documenting of the deviation is describing the conditions, as they currently exist, and the criteria, which represents what the user desires. The actual deviation will be the difference or gap between what is and what is desired. The statement of condition is uncovering and documenting the facts, as they exist. What is a fact? The statement of condition will of course depend on the nature and extent of the evidence or support that is examined and noted. For those facts, making up the statement of condition, the I/S professional will need to ensure that the information is accurate, well supported, and worded as clearly and precisely as possible. The statement of condition should document as many of the following attributes as appropriate of the problem. Activities Involved:- The specific business or administered activities that are being performed during Test Log generation are as follows: Procedures used to perform work. The specific step-by step activities that are utilized in producing the output from the identical activities. Outputs /Deliverables The products that are produced from the activity. Inputs - The triggers,events,or documents that cause this activity to be executed. Users/Customers served The organization ,individuvals,or class users/customers serviced by this activity.
- 77 -

Deficiencies noted The status of the results of executing this activity and any appropriate interpretation of those facts. The Criterion is the users statement of what is desired. It can be stated in the either negative or positive terms. For example , it could indicate the need to reduce the complaints or delays as well as desired processing turn around time. Work Paper to describe the problem, and document the statement of condition and the statement of criteria. For example the following Work paper provides the information for Test Log Documentation:

Field Requirements: Field Instructions for Entering Data Name of Software Tested : Put the name of the S/W or subsystem tested. Problem Description: Write a brief narrative description of the variance uncovered from expectations Statement of Conditions: Put the results of actual processing that occurred here. Statement of Criteria: Put what testers believe was the expected result from processing Effect of Deviation: If this can be estimated , testers should indicate what they believe the impact or effect of the problem will be on computer processing Cause of Problem: The testers should indicate what they believe is the cause of the problem, if known. If the testers re unable to do this , the work paper will be given to the development team and they should indicate the cause of the problem. Location of the Problem: The Tests should document where problem occurred as closely as possible. Recommended Action: The testers should indicate any recommended action they believe would be helpful to the project team. If not approved, the alternate action should be listed or the reason for not following the recommended action should be documented. Name of the S/W tested: Problem Description Statement of Condition Statement of Criteria Effect of Deviation Cause of a Problem Location of the Problem Recommended Action

13.2 Collecting Status Data

Four categories of data will be collected during testing. These are explained in the following paragraphs.

Test Results Data This data will include, Test factors -The factors incorporated in the plan, the validation of which becomes the Test Objective. Business objective The validation that specific business objectives have been met. Interface Objectives-Validation that data/Objects can be correctly passed among Software components. Functions/Sub functions-Identifiable Software components normally associated with the requirements of the software.
- 78 -

Units- The smallest identifiable software components Platform- The hardware and Software environment in which the software system will operate.
Test Transactions, Test Suites, and Test Events These are the test products produced by the test team to perform testing. Test transactions/events: The type of tests that will be conducted during the execution of tests, which will be based on software requirements. Inspections A verification of process deliverables against deliverable specifications. Reviews: Verification that the process deliverables / phases are meeting the users true needs.

This category includes a Description of the individual defects uncovered during the testing process. This description includes but not limited to : Data the defect uncovered Name of the Defect Location of the Defect Severity of the Defect Type of Defect How the defect was uncovered(Test Data/Test Script) The Test Logs should add to this information in the form of where the defect originated , when it was corrected, and when it was entered for retest.

Storing Data Collected during Testing

It is recommended that a database be established in which to store the results collected during testing. It is also suggested that the database be put in online through client/server systems so that with a vested interest in the status of the project can be readily accessed for the status update. As described the most common test Report is a simple Spread sheet , which indicates the project component for which the status is requested, the test that will be performed to determine the status of that component, and the results of testing at any point of time.

Developing Test Status Reports

Report Software Status Establish a Measurement Team Inventory Existing Project Measures Develop a Consistent Set of Project metrics Define Process Requirements Develop and Implement the Process Monitor the Process The Test process should produce a continuous series of reports that describe the status of testing. The test reports are for use of testers, test managers, and the software development team. The frequency of the test reports should be based on the discretion of the team and extensiveness of the test process.

Use of Function/Test matrix:

This shows which tests must be performed in order to validate the functions and also used to determine the status of testing. Many organizations use spreadsheet package to maintain test results. The intersection can be coded with a number or symbol to indicate the following: 1=Test is needed, but not performed 2=Test currently being performed - 79 -

3=MINOR DEFECT NOTED 4=Major defect noted 5=Test complete and function is defect free for the criteria included in this test TEST FUNCTION 1 2 3 4 5 6 7 8 A B C D E Function Test Matrix


Methods of Test Reporting

Reporting Tools - Use of word processing, database, defect tracking, and graphic tools to prepare test reports. Some Database test tools like Data Vision is a database reporting tool similar to Crystal Reports. Reports can be viewed and printed from the application or output as HTML, LaTeX2e, XML, DocBook, or tab- or comma-separated text files. From the LaTeX2e and DocBook output files you can in turn produce PDF, text, HTML, PostScript, and more. Some query tools available for Linux-based databases include: QMySQL dbMetrix PgAccess Cognos Powerhouse This is not yet available for Linux; Cognos is looking into what interest people have in the product to assess what their strategy should be with respect to the Linux ``market.'' GRG - GNU Report Generator The GRG program reads record and field information from a dBase3+ file, delimited ASCII text file or a SQL query to a RDBMS and produces a report listing. The program was loosely designed to produce TeX/LaTeX formatted output, but plain ASCII text, troff, PostScript, HTML or any other kind of ASCII based output format can be produced just as easily. Word Processing: One way of increasing the utility of computers and word processors for the teaching of writing may be to use software that will guide the processes of generating, organizing, composing and revising text. This allows each person to use the normal functions of the computer keyboard that are common to all word processors, email editors, order entry systems, and data base management products. From the Report Manager, however, you can quickly scan through any number of these reports and see how each person's history compares. A one-page summary report may be printed with either the Report Manager program or from the individual keyboard or keypad software at any time. Individual Reports include all of the following information. Status Report Word Processing Tests or Keypad Tests Basic Skills Tests or Data Entry Tests Progress Graph Game Scores Test Report for each test Test Director:

- 80 -

Facilitates consistent and repetitive testing process Central repository for all testing assets facilitates the adoption of a more consistent testing process, which can be repeated throughout the application life cycle Provides Analysis and Decision Support Graphs and reports help analyze application readiness at any point in the testing process Requirements coverage, run schedules, test execution progress, defect statistics can be used for production planning Provides Anytime, Anywhere access to Test Assets Using Test Directors web interface, tester, developers, business analysts and Client can participate and contribute to the testing process Traceability throughout the testing process Test Cases can be mapped to requirements providing adequate visibility over the test coverage of requirements Test Director links requirements to test cases and test cases to defects Manages Both Manual and Automated Testing Test Director can manage both manual and automated tests (Win Runner) Scheduling of automated tests can be effectively done using Test Director

Test Report Standards - Defining the components that should be included in a test report. Statistical Analysis - Ability to draw statistically valid conclusions from quantitative test results. Testing Data used for metrics

Testers are typically responsible for reporting their test status at regular intervals. The following measurements generated during testing are applicable: Total number of tests Number of Tests executed to date Number of tests executed successfully to date Data concerning software defects include Total number of defects corrected in each activity Total number of defects entered in each activity. Average duration between defect detection and defect correction Average effort to correct a defect Total number of defects remaining at delivery Software performance data us usually generated during system testing, once the software has been integrated and functional testing is complete. Average CPU utilization Average memory Utilization Measured I/O transaction rate
Test Reporting A final test report should be prepared at the conclusion of each test activity. This includes the following - 81 -

Individual Project Test Report Integration Test Report System Test Report Acceptance test Report

These test reports are designed to document the results of testing as defined in the testplan.The test report can be a combination of electronic data and hard copy. For example, if the function matrix is maintained electronically, there is no reason to print that, as paper report will summarize the data, draws appropriate conclusions and present recommendations.9 - Purpose of a Test Report: The test report has one immediate and three long term purposes. The immediate purpose is to provide information to customers of the software system so that they can determine whether the system is ready for production , and if so, to assess the potential consequences and initiate appropriate actions to minimize those consequences. The first of the three long term uses is for the project to trace problems in the event the application malfunctions in production. Knowing which functions have been correctly tested and which ones still contain defects can assist in taking corrective actions. The second long term purpose is to use the data to analyze the rework process for making changes to prevent the defects from occurring in the future. These defect prone components identify tasks/steps that if improved, could eliminate or minimize the occurrence of high frequency defects. The Third long term purpose is to show what was accomplished in case of an Y2K lawsuit. Individual Project Test Report These reports focus on the Individual projects(software system),when different testers should test individual projects, they should prepare a report on their results. Integration Test Report Integration testing tests the interfaces between individual projects. A good test plan will identify the interfaces and institute test conditions that will validate interfaces. Given is the Individual Project test report except that conditions tested are interfaces. 1.Scope of Test This section indicates which functions were and were not tested 2.Test Results This section indicates the results of testing, including any variance between what is and what should be 3.What works/What does not work - This section defines the functions that work and do not work and the interfaces that work and do not work 4. Recommendations This section recommends actions that should be taken to Fix functions /Interfaces that do not work. Make additional improvements System Test Reports A System Test plan standard that identified the objective of testing , what was to be tested, how was it to be tested, and when tests should occur. The system test Report should present the results of executing the test plan. If these details are maintained Electronically , then it need only be referenced , not included in the report. Acceptance Test Report There are two primary objectives of Acceptance testing Report .The first is to ensure that the system as implemented meets the real operating needs of the user/customer. If the defined requirements are those true needs, testing should have accomplished this objective. The second objective is to ensure that software system can operate in the real world user environment, which includes people skills and attitudes, time pressures, changing
- 82 -

business conditions, and so forth. The Acceptance Test Report should encompass these criterias for the User acceptance respectively.



The Test Logs obtained from the execution of the test results and finally the test reports should be designed to accomplish the following objectives: Provide Information to the customer whether the system should be placed into production, if so the potential consequences and appropriate actions to minimize these consequences. One Long term objective is for the Project and the other is for the information technology function. The project can use the test report to trace problems in the event the application malfunction in production. Knowing which functions have been correctly tested and which ones still contain defects can assist in taking corrective actions. The data can also be used to analyze the developmental process to make changes to prevent defects from occurring in the future. These defect prone components identify tasks/steps that if improved, could eliminate or minimize the occurrence of high frequency defects in future.

- 83 -

14 Test Report
A Test Report is a document that is prepared once the testing of a software product is complete and the delivery is to be made to the customer. This document would contain a summary of the entire project and would have to be presented in a way that any person who has not worked on the project would also get a good overview of the testing effort.

Contents of a Test Report

The contents of a test report are as follows: Executive Summary Overview Application Overview Testing Scope Test Details Test Approach Types of testing conducted Test Environment Tools Used Metrics Test Results Test Deliverables Recommendations These sections are explained as follows:

14.1 Executive Summary

This section would comprise of general information regarding the project, the client, the application, tools and people involved in such a way that it can be taken as a summary of the Test Report itself (i.e.) all the topics mentioned here would be elaborated in the various sections of the report. 1. Overview This comprises of 2 sections Application Overview and Testing Scope. Application Overview This would include detailed information on the application under test, the end users and a brief outline of the functionality as well. Testing Scope This would clearly outline the areas of the application that would / would not be tested by the QA team. This is done so that there would not be any misunderstandings between customer and QA as regards what needs to be tested and what does not need to be tested. This section would also contain information of Operating System / Browser combinations if Compatibility testing is included in the testing effort.
- 84 -

2. Test Details This section would contain the Test Approach, Types of Testing conducted, Test Environment and Tools Used. Test Approach This would discuss the strategy followed for executing the project. This could include information on how coordination was achieved between Onsite and Offshore teams, any innovative methods used for automation or for reducing repetitive workload on the testers, how information and daily / weekly deliverables were delivered to the client etc. Types of testing conducted This section would mention any specific types of testing performed (i.e.) Functional, Compatibility, Performance, Usability etc along with related specifications. Test Environment This would contain information on the Hardware and Software requirements for the project (i.e.) server configuration, client machine configuration, specific software installations required etc. Tools used This section would include information on any tools that were used for testing the project. They could be functional or performance testing automation tools, defect management tools, project tracking tools or any other tools which made the testing work easier. 3. Metrics This section would include details on total number of test cases executed in the course of the project, number of defects found etc. Calculations like defects found per test case or number of test cases executed per day per person etc would also be entered in this section. This can be used in calculating the efficiency of the testing effort. 4. Test Results This section is similar to the Metrics section, but is more for showcasing the salient features of the testing effort. Incase many defects have been logged for the project, graphs can be generated accordingly and depicted in this section. The graphs can be for Defects per build, Defects based on severity, Defects based on Status (i.e.) how many were fixed and how many rejected etc.

5. Test Deliverables This section would include links to the various documents prepared in the course of the testing project (i.e.) Test Plan, Test Procedures, Test Logs, Release Report etc. 6. Recommendations
- 85 -

This section would include any recommendations from the QA team to the client on the product tested. It could also mention the list of known defects which have been logged by QA but not yet fixed by the development team so that they can be taken care of in the next release of the application.

- 86 -

15 Defect Management
15.1 Defect
A mismatch in the application and its specification is a defect. A software error is present when the program does not do what its end user expects it to do.

15.2 Defect Fundamentals

A Defect is a product anomaly or flaw. Defects include such things as omissions and imperfections found during testing phases. Symptoms (flaws) of faults contained in software that is sufficiently mature for production will be considered as defects. Deviations from expectation that is to be tracked and resolved is also termed a defect. An evaluation of defects discovered during testing provides the best indication of software quality. Quality is the indication of how well the system meets the requirements. So in this context defects are identified as any failure to meet the system requirements. Defect evaluation is based on methods that range from simple number count to rigorous statistical modeling. Rigorous evaluation uses assumptions about the arrival or discovery rates of defects during the testing process. The actual data about defect rates are then fit to the model. Such an evaluation estimates the current system reliability and predicts how the reliability will grow if testing and defect removal continue. This evaluation is described as system reliability growth modelling

- 87 -


Defect Life Cycle

15.3 Defect Tracking

After a defect has been found, it must be reported to development so that it can be fixed.

The Initial State of a defect will be New. The Project Lead of the development team will review the defect and set it to one of the following statuses: Open Accepts the bug and assigns it to a developer. Invalid Bug The reported bug is not valid one as per the requirements/design As Designed This is an intended functionality as per the requirements/design Deferred This will be an enhancement. Duplicate The bug has already been reported.
- 88 -

Document Once it is set to any of the above statuses apart from Open, and the testing team does not agree with the development team it is set to document status.

Once the development team has started working on the defect the status is set to WIP ((Work in Progress) or if the development team is waiting for a go ahead or some technical feedback, they will set to Dev Waiting After the development team has fixed the defect, the status is set to FIXED, which means the defect is ready to re-test. On re-testing the defect, and the defect still exists, the status is set to REOPENED, which will follow the same cycle as an open defect. If the fixed defect satisfies the requirements/passes the test case, it is set to Closed.

15.4 Defect Classification

The severity of bugs will be classified as follows: Critical High The problem prevents further processing and testing. The Development Team must be informed immediately and they need to take corrective action immediately. The problem affects selected processing to a significant degree, making it inoperable, Cause data loss, or could cause a user to make an incorrect decision or entry. The Development Team must be informed that day, and they need to take corrective action within 0 24 hours. The problem affects selected processing, but has a work-around that allows continued processing and testing. No data loss is suffered. These may be cosmetic problems that hamper usability or divulge client-specific information. The Development Team must be informed within 24 hours, and they need to take corrective action within 24 - 48 hours. The problem is cosmetic, and/or does not affect further processing and testing. The Development Team must be informed within 48 hours, and they need to take corrective action within 48 - 96 hours.



15.5 Defect Reporting Guidelines

- 89 -

The key to making a good report is providing the development staff with as much information as necessary to reproduce the bug. This can be broken down into 5 points: 1) Give a brief description of the problem 2) List the steps that are needed to reproduce the bug or problem 3) Supply all relevant information such as version, project and data used. 4) Supply a copy of all relevant reports and data including copies of the expected results. 5) Summarize what you think the problem is. When you are reporting a defect the more information you supply, the easier it will be for the developers to determine the problem and fix it. Simple problems can have a simple report, but the more complex the problem the more information the developer is going to need. For example: cosmetic errors may only require a brief description of the screen, how to get it and what needs to be changed. However, an error in processing will require a more detailed description, such as: 1) The name of the process and how to get to it. 2) Documentation on what was expected. (Expected results) 3) The source of the expected results, if available. This includes spread sheets, an earlier version of the software and any formulas used) 4) Documentation on what actually happened. (Perceived results) 5) An explanation of how the results differed. 6) Identify the individual items that are wrong. 7) If specific data is involved, a copy of the data both before and after the process should be included. 8) Copies of any output should be included. As a rule the detail of your report will increase based on a) the severity of the bug, b) the level of the processing, c) the complexity of reproducing the bug. Anatomy of a bug report Bug reports need to do more than just describe the bug. They have to give developers something to work with so that they can successfully reproduce the problem.
- 90 -

In most cases the more information correct information given the better. The report should explain exactly how to reproduce the problem and an explanation of exactly what the problem is. The basic items in a report are as follows: Version: This is very important. In most cases the product is not static, developers will have been working on it and if theyve found a bug it may already have been reported or even fixed. In either case, they need to know which version to use when testing out the bug. If you are developing more than one product Identify the product in question. Unless you are reporting something very simple, such as a cosmetic error on a screen, you should include a dataset that exhibits the error. If youre reporting a processing error, you should include two versions of the dataset, one before the process and one after. If the dataset from before the process is not included, developers will be forced to try and find the bug based on forensic evidence. With the data, developers can trace what is happening. Steps: List the steps taken to recreate the bug. Include all proper menu names, dont abbreviate and dont assume anything. After youve finished writing down the steps, follow them - make sure youve included everything you type and do to get to the problem. If there are parameters, list them. If you have to enter any data, supply the exact data entered. Go through the process again and see if there are any steps that can be removed. When you report the steps they should be the clearest steps to recreating the bug. Description: Explain what is wrong - Try to weed out any extraneous information, but detail what is wrong. Include a list of what was expected. Remember report one problem at a time, dont combine bugs in one report.
- 91 -

Product: Data:

Supporting documentation: If available, supply documentation. If the process is a report, include a copy of the report with the problem areas highlighted. Include what you expected. If you have a report to compare against, include it and its source information (if its a printout from a previous version, include the version number and the dataset used) This information should be stored in a centralized location so that Developers and Testers have access to the information. The developers need it to reproduce the bug, identify it and fix it. Testers will need this information for later regression testing and verification.



A bug report is a case against a product. In order to work it must supply all necessary information to not only identify the problem but what is needed to fix it as well. It is not enough to say that something is wrong. The report must also say what the system should be doing. The report should be written in clear concise steps, so that someone who has never seen the system can follow the steps and reproduce the problem. It should include information about the product, including the version number, what data was used. The more organized information provided the better the report will be.

- 92 -

16 Automation
What is Automation Automated testing is automating the manual testing process currently in use

16.1 Why Automate the Testing Process?

Today, rigorous application testing is a critical part of virtually all software development projects. As more organizations develop mission-critical systems to support their business activities, the need is greatly increased for testing methods that support business objectives. It is necessary to ensure that these systems are reliable, built according to specification, and have the ability to support business processes. Many internal and external factors are forcing organizations to ensure a high level of software quality and reliability. In the past, most software tests were performed using manual methods. This required a large staff of test personnel to perform expensive, and time-consuming manual test procedures. Owing to the size and complexity of todays advanced software applications, manual testing is no longer a viable option for most testing situations. Every organization has unique reasons for automating software quality activities, but several reasons are common across industries.

Using Testing Effectively

By definition, testing is a repetitive activity. The very nature of application software development dictates that no matter which methods are employed to carry out testing (manual or automated), they remain repetitious throughout the development lifecycle. Automation of testing processes allows machines to complete the tedious, repetitive work while human personnel perform other tasks. Automation allows the tester to reduce or eliminate the required think time or read time necessary for the manual interpretation of when or where to click the mouse or press the enter key. An automated test executes the next operation in the test hierarchy at machine speed, allowing tests to be completed many times faster than the fastest individual. Furthermore, some types of testing, such as load/stress testing, are virtually impossible to perform manually.

Reducing Testing Costs

The cost of performing manual testing is prohibitive when compared to automated methods. The reason is that computers can execute instructions many times faster, and with fewer errors than individuals. Many automated testing tools can replicate the activity of a large number of users (and their associated transactions) using a single computer. Therefore, load/stress testing using - 93 -

automated methods require only a fraction of the computer hardware that would be necessary to complete a manual test. Imagine performing a load test on a typical distributed client/server application on which 50 concurrent users were planned. To do the testing manually, 50 application users employing 50 PCs with associated software, an available network, and a cadre of coordinators to relay instructions to the users would be required. With an automated scenario, the entire test operation could be created on a single machine having the ability to run and rerun the test as necessary, at night or on weekends without having to assemble an army of end users. As another example, imagine the same application used by hundreds or thousands of users. It is easy to see why manual methods for load/stress testing is an expensive and logistical nightmare.

Replicating Testing Across Different Platforms

Automation allows the testing organization to perform consistent and repeatable tests. When applications need to be deployed across different hardware or software platforms, standard or benchmark tests can be created and repeated on target platforms to ensure that new platforms operate consistently.

Repeatability and Control

By using automated techniques, the tester has a very high degree of control over which types of tests are being performed, and how the tests will be executed. Using automated tests enforces consistent procedures that allow developers to evaluate the effect of various application modifications as well as the effect of various user actions. For example, automated tests can be built that extract variable data from external files or applications and then run a test using the data as an input value. Most importantly, automated tests can be executed as many times as necessary without requiring a user to recreate a test script each time the test is run.

Greater Application Coverage

The productivity gains delivered by automated testing allow and encourage organizations to test more often and more completely. Greater application test coverage also reduces the risk of exposing users to malfunctioning or non-compliant software. In some industries such as healthcare and pharmaceuticals, organizations are required to comply with strict quality regulations as well as being required to document their quality assurance efforts for all parts of their systems.

- 94 -

16.2 Automation Life Cycle

Identifying Tests Requiring Automation

Most, but not all, types of tests can be automated. Certain types of tests like user comprehension tests, tests that run only once, and tests that require constant human intervention are usually not worth the investment to automate. The following are examples of criteria that can be used to identify tests that are prime candidates for automation. High Path Frequency - Automated testing can be used to verify the performance of application paths that are used with a high degree of frequency when the software is running in full production. Examples include: creating customer records, invoicing and other high volume activities where software failures would occur frequently. Critical Business Processes - In many situations, software applications can literally define or control the core of a companys business. If the application fails, the company can face extreme disruptions in critical operations. Mission-critical processes are prime candidates for automated testing. Examples include: financial month-end closings, production planning, sales order entry and other core activities. Any application with a high-degree of risk associated with a failure is a good candidate for test automation.
- 95 -

Repetitive Testing - If a testing procedure can be reused many times, it is also a prime candidate for automation. For example, common outline files can be created to establish a testing session, close a testing session and apply testing values. These automated modules can be used again and again without having to rebuild the test scripts. This modular approach saves time and money when compared to creating a new end-to-end script for each and every test. Applications with a Long Life Span - If an application is planned to be in production for a long period of time, the greater the benefits are from automation. What to Look For in a Testing Tool Choosing an automated software testing tool is an important step, and one which often poses enterprise-wide implications. Here are several key issues, which should be addressed when selecting an application testing solution.

Test Planning and Management

A robust testing tool should have the capability to manage the testing process, provide organization for testing components, and create meaningful end-user and management reports. It should also allow users to include non-automated testing procedures within automated test plans and test results. A robust tool will allow users to integrate existing test results into an automated test plan. Finally, an automated test should be able to link business requirements to test results, allowing users to evaluate application readiness based upon the application's ability to support the business requirements.

Testing Product Integration

Testing tools should provide tightly integrated modules that support test component reusability. Test components built for performing functional tests should also support other types of testing including regression and load/stress testing. All products within the testing product environment should be based upon a common, easy-to-understand language. User training and experience gained in performing one testing task should be transferable to other testing tasks. Also, the architecture of the testing tool environment should be open to support interaction with other technologies such as defect or bug tracking packages.

Internet/Intranet Testing
A good tool will have the ability to support testing within the scope of a web browser. The tests created for testing Internet or intranet-based applications should be portable across browsers, and should automatically adjust for different load times and performance levels.

Ease of Use

Testing tools should be engineered to be usable by non-programmers and application end-users. With much of the testing responsibility shifting from the development staff to the departmental level, a testing tool that requires programming skills is unusable by most organizations. Even if programmers are responsible for testing, the testing tool itself should have a short learning curve.

GUI and Client/Server Testing

A robust testing tool should support testing with a variety of user interfaces and create simple-to manage, easy-to-modify tests. Test component reusability should be a cornerstone of the product - 96 -


Load and Performance Testing

The selected testing solution should allow users to perform meaningful load and performance tests to accurately measure system performance. It should also provide test results in an easy-to-understand reporting format.

16.3 Preparing the Test Environment

Once the test cases have been created, the test environment can be prepared. The test environment is defined as the complete set of steps necessary to execute the test as described in the test plan. The test environment includes initial set up and description of the environment, and the procedures needed for installation and restoration of the environment. Description - Document the technical environment needed to execute the tests. Test Schedule - Identify the times during which your testing facilities will be used for a given test. Make sure that other groups that might share these resources are informed of this schedule. Operational Support - Identify any support needed from other parts of your organization. Installation Procedures - Outline the procedures necessary to install the application software to be tested. Restoration Procedures - Finally, outline those procedures needed to restore the test environment to its original state. By doing this, you are ready to re-execute tests or prepare for a different set of tests. Inputs to the Test Environment Preparation Process Technical Environment Descriptions Approved Test Plan Test Execution Schedules Resource Allocation Schedule Application Software to be installed

- 97 -

Test Planning
Careful planning is the key to any successful process. To guarantee the best possible result from an automated testing program, those evaluating test automation should consider these fundamental planning steps. The time invested in detailed planning significantly improves the benefits resulting from test automation.

Evaluating Business Requirements

Begin the automated testing process by defining exactly what tasks your application software should accomplish in terms of the actual business activities of the end-user. The definition of these tasks, or business requirements, defines the high-level, functional requirements of the software system in question. These business requirements should be defined in such a way as to make it abundantly clear that the software system correctly (or incorrectly) performs the necessary business functions. For example, a business requirement for a payroll application might be to calculate a salary, or to print a salary check.

Creating a Test Plan

For the greatest return on automated testing, a testing plan should be created at the same time the software application requirements are defined. This enables the testing team to define the tests, locate and configure test-related hardware and software products and coordinate the human resources required to complete all testing. This plan is very much a living document that should evolve as the application functions become more clearly defined. A good testing plan should be reviewed and approved by the test team, the software development team, all user groups and the organizations management. The following items detail the input and output components of the test planning process.

Inputs to the Test Planning Process

Application Requirements - What is the application intended to do? These should be stated in the terms of the business requirements of the end users. Application Implementation Schedules - When is the scheduled release? When are updates or enhancements planned? Are there any specific events or actions that are dependent upon the application? Acceptance Criteria for implementation - What critical actions must the application accomplish before it can be deployed? This information forms the basis for making informed decisions on whether or not the application is ready to deploy.

Test Design and Development

After the test components have been defined, the standardized test cases can be created that will be used to test the application. The type and number of test cases needed will be dictated by the testing plan.
Performance Testing Process & Methodology 98 Proprietary & Confidential -

A test case identifies the specific input values that will be sent to the application, the procedures for applying those inputs, and the expected application values for the procedure being tested. A proper test case will include the following key components: Test Case Name(s) - Each test case must have a unique name, so that the results of these test elements can be traced and analyzed. Test Case Prerequisites - Identify set up or testing criteria that must be established before a test can be successfully executed. Test Case Execution Order - Specify any relationships, run orders and dependencies that might exist between test cases. Test Procedures Identify the application steps necessary to complete the test case. Input Values - This section of the test case identifies the values to be supplied to the application as input including, if necessary, the action to be completed. Expected Results - Document all screen identifier(s) and expected value(s) that must be verified as part of the test. These expected results will be used to measure the acceptance criteria, and therefore the ultimate success of the test. Test Data Sources - Take note of the sources for extracting test data if it is not included in the test case. Inputs to the Test Design and Construction Process Test Case Documentation Standards Test Case Naming Standards Approved Test Plan Business Process Documentation Business Process Flow Test Data sources Outputs from the Test Design and Construction Process Revised Test Plan Test Procedures for each Test Case Test Case(s) for each application function described in the test plan Procedures for test set up, test execution and restoration

Executing the Test

The test is now ready to be run. This step applies the test cases identified by the test plan, documents the results, and validates those results against expected performance. Specific performance measurements of the test execution phase include: Application of Test Cases The test cases previously created are applied to the target software application as described in the testing environment Documentation - Activities within the test execution are logged and analyzed as follows: Actual Results achieved during test execution are compared to expected application behavior from the test cases Test Case completion status (Pass/Fail) Actual results of the behavior of the technical test environment Deviations taken from the test plan or test process
Performance Testing Process & Methodology 99 Proprietary & Confidential -

Inputs to the Test Execution Process Approved Test Plan Documented Test Cases Stabilized, repeatable, test execution environment Standardized Test Logging Procedures Outputs from the Test Execution Process Test Execution Log(s) Restored test environment The test execution phase of your software test process will control how the test gets applied to the application. This step of the process can range from very chaotic to very simple and schedule driven. The problems experienced in test execution are usually attributed to not properly performing steps from earlier in the process. Additionally, there may be several test execution cycles necessary to complete all the necessary types of testing required for your application. For example, a test execution may be required for the functional testing of an application, and a separate test execution cycle may be required for the stress/volume testing of the same application. A complete and thorough test plan will identify this need and many of the test cases can be used for both test cycles. The secret to a controlled test execution is comprehensive planning. Without an adequate test plan in place to control your entire test process, you may inadvertently cause problems for subsequent testing.

Measuring the Results

This step evaluates the results of the test as compared to the acceptance criteria set down in the test plan. Specific elements to be measured and analyzed include: Test Execution Log Review - The Log Review compiles a listing of the activities of all test cases, noting those that passed, failed or were not executed. Determine Application Status - This step identifies the overall status of the application after testing, for example: ready for release, needs more testing, etc. Test Execution Statistics - This summary identifies the total number of tests that were executed, the type of test, and the completion status. Application Defects - This final and very important report identifies potential defects in the software, including application processes that need to be analyzed further.

16.4 Automation Methods

Capture/Playback Approach The Capture/Playback tools capture the sequence of
manual operations in a test script that are entered by the test engineer. These sequences are played back during the test execution. The benefit of this approach is that the captured session can be re-run at some later point in time to ensure that the system performs the required behavior. The short-comings of Capture/Playback are that in many cases, if the system functionality changes, the capture/playback session will need to be completely re-run to capture the new sequence of user interactions. Tools like WinRunner provide a scripting language, and it is possible for engineers to edit and maintain such scripts. This sometimes reduces the effort over the completely manual approach, however overall savings is usually minimal.
Performance Testing Process & Methodology 100 Proprietary & Confidential -

Data Driven Approach

Data driven approach is a test that plays back the same user actions but with varying input values. This allows one script to test multiple sets of positive data. This is applicable when large volumes and different sets of data need to be fed to the application and tested for correctness. The benefit of this approach is that the time consumed is less and accurate than manually testing it. Testing can be done with both positive and negative approach simultaneously.

Test Script execution:

In this phase we execute the scripts that are already created. Scripts need to be reviewed and validated for results and accepted as functioning as expected before they are used live. Steps to be followed before execution of scripts: 1.Test tool to be installed in the machine. 2. Test environment /application to be tested to be installed in the machine. 3. Prerequisite for running the scripts such as tool settings, playback options, necessary data table or data pool updation needs to be taken care. 4.Select the script that needs to be executed and run it 5.Wait until execution is done. 6.Analysis the results via Test manager or in the logs.

Test script execution process:

Performance Testing Process & Methodology 101 -

Proprietary & Confidential

Test tool ready

Test ready


Tool settings, Playback options Script execution

Result analysis

Defect management

Performance Testing Process & Methodology 102 -

Proprietary & Confidential

17 General automation tool comparison

Anyone who has contemplated the implementation of an automated test tool has quickly realized the wide variety of options on the market in terms of both the kinds of test tools being offered and the number of vendors. The best tool for any particular situation depends on the system engineering environment that applies and the testing methodology that will be used, which in turn will dictate how automation will be invoked to support the process. This appendix evaluates major tool vendors on their test tool characteristics, test execution capability, tool integration capability, test reporting capability, performance testing and analysis, and vendor qualification. The following tool vendors evaluated are Compuware, Empirix/RSW, Mercury, Rational, and Segue.

17.1 Functional Test Tool Matrix

The Tool Matrix is provided for quick and easy reference to the capabilities of the test tools. Each category in the matrix is given a rating of 1 5. 1 = Excellent support for this functionality, 2 = Good support but lacking or another tool provides more effective support, 3 = Basic/ support only. 4 = This is only supported by use of an API call or third party add-in but not included in the general test tool/below average, 5 = No support. In general a set of criteria can be built up by using this matrix and an indicative score obtained to help in the evaluation process. Usually the lower the score the better but this is subjective and is based on the experience of the author and the test professionals opinions used to create this document. A detailed description is given below of each of the categories used in the matrix.

17.2 Record and Playback

This category details how easy it is to record & playback a test. Does the tool support low-level recording (mouse drags, exact screen location)? Is there object recognition when recording and playing back or does it appear to record ok but then on playback (without environment change or unique ids, etc changes) fail? How easy is it to read the recorded script. When automating, this is the first thing that most test professionals will do. They will record a simple script; look at the code and then playback. This is very similar to recording a macro in say Microsoft Access. Eventually record and playback becomes less and less part of the automation process as it is usually more robust to use the built-in functions to directly test objects, databases, etc. However this should be done as a minimum in the evaluation process because if the tool of choice cannot recognize the applications objects then the automation process will be a very tedious experience.

Performance Testing Process & Methodology 103 -

Proprietary & Confidential


Web Testing

Web based functionality on most applications is now a part of everyday life. As such the test tool should provide good web based test functionality in addition to its client/server functions. In judging the rating for this category I looked at the tools native support for HTML tables, frames, DOM, various platforms for browsers, Web site maps and links. Web testing can be riddled with problems if various considerations are not taken into account. Here are a few examples Are there functions to tell me when the page has finished loading? Can I tell the test tool to wait until an image appears? Can I test whether links are valid or not? Can I test web based objects functions like is it enabled, does it contain data, etc. Are there facilities that will allow me to programmatically look for objects of a certain type on a web page or locate a specific object? Can I extract data from the web page itself? E.g. the title? A hidden form element? With Client server testing the target customer is usually well defined you know what network operating system you will be using, the applications and so on but on the web it is far different. A person may be connecting from the USA or Africa, they may be disabled, they may use various browsers, and the screen resolution on their computer will be different. They will speak different languages, will have fast connections and slow connections, connect using MAC, Linux or Windows, etc, etc. So the cost to set up a test environment is usually greater than for a client server test where the environment is fairly well defined.

17.4 Database Tests

Most applications will provide the facility to preserve data outside of itself. This is usually achieved by holding the data in a Database. As such, checking what is in the backend database usually verifies the proper validation of tests carried out on the front end of an application. Because of the many databases available e.g. Oracle, DB2, SQLServer, Sybase, Informix, Ingres, etc all of them support a universal query language known as SQL and a protocol for communicating with these databases called ODBC (JDBC can be used on java environments). I have looked at all the tools support for SQL, ODBC and how they hold returned data e.g. is this in an array, a cursor, a variable, etc. How does the tool manipulate this returned data? Can it call stored procedures and supply required input variables? What is the range of functions supplied for this testing?

17.5 Data Functions

As mentioned above applications usually provide a facility for storing data off line. So to test this, we will need to create data to input into the application. I have looked at
Performance Testing Process & Methodology 104 Proprietary & Confidential -

all the tools facilities for creating and manipulating data. Does the tool allow you to specify the type of data you want? Can you automatically generate data? Can you interface with files, spreadsheets, etc to create, extract data? Can you randomise the access to that data? Is the data access truly random? This functionality is normally more important than database tests as the databases will usually have their own interface for running queries. However applications (except for manual input) do not usually provide facilities for bulk data input. The added benefit (as I have found) is this functionality can be used for a production reason e.g. for the aforementioned bulk data input sometimes carried out in data migration or application upgrades. These functions are also very important as you move from the record/playback phase, to data-driven to framework testing. Data-driven tests are tests that replace hard coded names, address, numbers; etc with variables supplied from an external source usually a CSV (Comma Separated variable) file, spreadsheet or database. Frameworks are usually the ultimate goal in deploying automation test tools. Frameworks provide an interface to all the applications under test by exposing a suitable list of functions, databases, etc. This allows an inexperienced tester/user to run tests by just running/providing the test framework with know commands/variables. A test framework has parallels to Software frameworks where you develop an encapsulation layer of software (framework) around the applications, databases etc and expose functions, classes, methods etc that is used to call the underlying applications, return data, input data, etc. However to do this requires a lot of time, skilled resources and money to facilitate the first two.

17.6 Object Mapping

If you are in a role that can help influence the design of a product, try to get the development/design team to use standard and not custom objects. Then hopefully you will not need this functionality. However you may find that most (hopefully) of the application has been implemented using standard objects supported by your test tool vendor but there may be a few objects that are custom ones. Most custom objects will behave like a similar standard control here are a few standard objects that are seen in everyday applications.

Pushbuttons Checkboxes Radio buttons List views Edit boxes Combo boxes

If you have a custom object that behaves like one of these are you able to map (tell the test tool that the custom control behaves like the standard) control? Does it support all the standard controls methods? Can you add the custom control to its own class of control?

Performance Testing Process & Methodology 105 -

Proprietary & Confidential

17.7 Image Testing

Lets hope this is not a major part of your testing effort but occasionally you may have to use this to test bit map and similar images. Also when the application has painted controls like those in the calculator app found on a lot of windows applications you may need to use this. At least one of the tools allows you to map painted controls to standard controls but to do this you have to rely on the screen co-ordinates of the image. Does the tool provide OCR (optical character recognition)? Can it compare one image against another? How fast does the compare take? If the compare fails how long does that take? Does the tool allow you to mask certain areas of the screen when comparing. I have looked at these facilities in the base tool set.

17.8 Test/Error recovery

This can be one of the most difficult areas to automate but if it is automated, it provides the foundation to produce a truly robust test suite. Suppose the application crashes while I am testing what can I do? If a function does not receive the correct information how can I handle this? If I get an error message how do I deal with that? If I access a web site and get a warning what do I do? I cannot get a database connection how do I skip those tests? The test tool should provide facilities to handle the above questions. I looked at built in wizards of the test tools for standard test recovery (when you finish tests or when a script fails). Error recovery caused by the application and environment. How easy is it to build this into your code? The rating given will depend on how much errors the tool can capture, the types of errors, how it recovers from errors, etc.

17.9 Object Name Map

As you test your application using the test tool of your choice you will notice that it records actions against the objects that it interacts with. These objects are either identified through the co-ordinates on the screen or preferably via some unique object reference referred to as a tag, object ID, index, name, etc. Firstly the tool should provide services to uniquely identify each object it interacts with and by various means. The last and least desirable should be by co-ordinates on the screen. Once you are well into automation and build up 10s and 100s of scripts that reference these objects you will want to have a mechanism that provides an easy update if the application being tested changes. All tools provide a search and replace facility but the best implementations are those that provide a central repository to store these object identities. The premise is it is better to change the reference in one place rather than having to go through each of the scripts to replace it there. I found this to be true but not as big a point as some have stated because those tools that dont support the central repository scheme; can be programmed to reference windows and object names in one place (say via a variable) and that variable can be used throughout the script (where that object appears). Does the Object Name Map allow you to alias the name or change the name given by the tool to some more meaningful name?
Performance Testing Process & Methodology 106 Proprietary & Confidential -

17.10Object Identity Tool

Once you become more proficient with automation testing one of the primary means of identifying objects will be via an ID Tool. A sort of spy that looks at the internals of the object giving you details like the object name, ID and similar. This will allow you to reference that object within a function call. The tool should give you details of some of the objects properties, especially those associated with uniquely identifying the object or window. The tool will usually provide the tester with a point and ID service where you can use the mouse to point at the object and in some window you will see all of that objects IDs and properties. A lot of the tools will allow you to search all the open applications in one swoop and show you the result in a tree that you can look at when required.

17.11Extensible Language
Here is a question that you will here time and time again in automation forums. How do I get {insert test tool name here} to do such and such, there will be one of four answers. I dont know It cant do it It can do it using the function x, y or Z It cant in the standard language but you can do it like this

What we are concerned with in this section is the last answer e.g. if the standard test language does not support it can I create a DLL or extend the language in some way to do it? This is usually an advanced topic and is not encountered until the trained tester has been using the tool for at least 6 12 months. However when this is encountered the tool should support language extension. If via DLLs then the tester must have knowledge of a traditional development language e.g. C, C++ or VB. For instance if I wanted to extend a tool that could use DLLs created by VB I would need to have Visual Basic then open say an ActiveX dll project, create a class containing various methods (similar to functions) then I would make a dll file. Register it on the machine then reference that dll from the test tool calling the methods according to their specification. This will sound a lot clearer as you go on in the tools and this document will be updated to include advanced topics like this in extending the tools capabilities. Some tools provide extension by allowing you to create user defined functions, methods, classes, etc but these are normally a mixture of the already supported data types, functions, etc rather than extending the tool beyond its released functionality. Because this is an advanced topic I have not taken into account ease of use, as those people who have got to this level should have already exhausted the current capabilities of the tools. So want to use external functions like win32api functions and so on and should have a good grasp of programming.

Performance Testing Process & Methodology 107 -

Proprietary & Confidential

17.12Environment Support
How many environments does the tool support out the box? Does it support the latest Java release, what Oracle, Powerbuilder, WAP, etc. Most tools can interface to unsupported environments if the developers in that environment provide classes, dlls etc that expose some of the applications details but whether a developer will or has time to do this is another question. Ultimately this is the most important part of automation. Environment support. If the tool does not support your environment/application then you are in trouble and in most cases you will need to revert to manually testing the application (more shelf ware).

How well does the tool integrate with other tools. This is becoming more and more important. Does the tool allow you to run it from various test management suites? Can you raise a bug directly from the tool and feed the information gathered from your test logs into it? Does it integrate with products like word, excel or requirements management tools? When managing large test projects with an automation team greater than five and testers totaling more than ten. The management aspect and the tools integration moves further up the importance ladder. An example could be a major Bank wants to redesign its workflow management system to allow faster processing of customer queries. The anticipated requirements for the new workflow software numbers in the thousands. To test these requirements 40,000 test cases have been identified 20,000 of these can be automated. How do I manage this? This is where a test management tool comes in real handy. Also how do I manage the bugs raised as a result of automation testing, etc? Integration becomes very important rather than having separate systems that dont share data that may require duplication of information. The companies that will score larger on these are those that provide tools outside the testing arena as they can build in integration to their other products and so when it comes down to the wire on some projects, we have gone with the tool that integrated with the products we already had.

In my opinion cost is the least significant in this matrix, why? Because all the tools are similar in price except Visual Test that is at least 5 times cheaper than the rest but as you will see from the matrix there is a reason. Although very functional it does not provide the range of facilities that the other tools do. Price typically ranges from $2,900 - $5,000 (depending on quantity brought, packages, etc) in the US and around 2,900 - 5,000 in the UK for the base tools included in this document. So you know the tools will all cost a similar price it is usually a case of which one will do the job for me rather than which is the cheapest. Visual Test I believe will prove to be a bigger hit as it expands its functional range it was not that long ago where it did not support web based testing.
Performance Testing Process & Methodology 108 Proprietary & Confidential -

The prices are kept this high because they can. All the tools are roughly the same price and the volumes of sales is low relative to say a fully blown programming language IDE like JBuilder or Visual C++ which are a lot more function rich and flexible than any of the test tools. On top of the above prices you usually pay an additional maintenance fee of between 10 and 20%. There are not many applications I know that cost this much per license not even some very advanced operating systems. However it is all a matter of supply. The bigger the supply the less the price as you can spread the development costs more. However I do not anticipate a move on the prices upwards as this seems to be the price the market will tolerate. Visual Test also provides a free runtime license.

17.15Ease Of Use
This section is very subjective but I have used testers (my guinea pigs) of various levels and got them from scratch to use each of the tools. In more cases than not they have agreed on which was the easiest to use (initially). Obviously this can change as the tester becomes more experienced and the issues of say extensibility, script maintenance, integration, data-driven tests, etc are required. However this score is based on the productivity that can be gained in say the first three months when those issues are not such a big concern. Ease of use includes out the box functions, debugging facilities, layout on screen, help files and user manuals.

In the UK this can be a problem as most of the test tool vendors are based in the USA with satellite branches in the UK. Just from my own experience and the testers I know in the UK. We have found Mercury to be the best for support, then Compuware, Rational and last Segue. However having said that you can find a lot of resources for Segue on the Internet including a forum at that can provide most of the answers rather than ringing the support line. On their website Segue and Mercury provide many useful user and vendor contributed material. I have also included various other criteria like the availability of skilled resources, online resources, validity of responses from the helpdesk, speed of responses and similar

17.17Object Tests
Now presuming the tool of choice does work with the application you wish to test what services does it provide for testing object properties? Can it validate several properties at once? Can it validate several objects at once? Can you set object properties to capture the application state? This should form the bulk of your verification as far as the automation process is concerned so I have looked at the tools facilities on client/server as well as web based applications.
Performance Testing Process & Methodology 109 Proprietary & Confidential -

What will follow after the matrix is a tool-by-tool comparison under the appropriate heading (as listed above) so that the user can get a feel for the tools functionality side by side. Each category in the matrix is given a rating of 1 5. 1 = Excellent support for this functionality, 2 = Good support but lacking or another tool provides more effective support, 3 = Basic/ support only. 4 = This is only supported by use of an API call or third party add-in but not included in the general test tool/below average, 5 = No support. R e c o r d & P l a y b a c k W e b T e s t i n g D a t a b a s e t e s t s D a t a f u n c t i o n s O b j e c t M a p p i n g I m a g e t e s t i n g T e s t / E r r o r r e c o v e r y 2 2 1 2 2 1 2 1 4 4 O b j e c t N a m e M a p O b j e c t I d e n t i t y T o o l 2 1 2 1 1 2 2 1 2 1 E x t e n s i b l e L a n g u a g e 1 2 2 3 2 E n v i r o n m e n t s u p p o r t 1 1 3 2 1 3 2 3 1 2 2 2 3 3 1 1 2 2 2 2 1 1 1 2 1 I n t e g r a t i o n C o s t E a s e o f u s e S u p p o r t O b j e c t T e s t s

WinRunner QA Run Silk Test Visual Test Robot

2 1 1 3 1

1 2 2 3 2

1 1 1 4 1

2 2 2 3 1

1 1 1 2 1

1 1 1 2 1

17.19Matrix score
Win Runner = 24 QARun = 25 SilkTest = 24 Visual Test = 39
Proprietary & Confidential -

Performance Testing Process & Methodology 110 -

Robot = 24

Performance Testing Process & Methodology 111 -

Proprietary & Confidential

18 Sample Test Automation Tool

Rational offers the most complete lifecycle toolset (including testing) of these vendors for the windows platform. When it comes to Object Oriented development they are the acknowledged leaders with most of the leading OO experts working for them. Some of their products are worldwide leaders e.g. Rational Robot, Rational Rose, Clear case, RequistePro, etc.Their Unified Process is a very good development model that I have been involved with which allows mapping of requirements to use cases, test cases and a whole set of tools to support the process.

18.1 Rational Suite of tools

Rational RequisitePro is a requirements management tool that helps project teams control the development process. RequisitePro organizes your requirements by linking Microsoft Word to a requirements repository and providing traceability and change management throughout the project lifecycle. A baseline version of RequisitePro is included with Rational Test Manager. When you define a test requirement in RequisitePro, you can access it in Test Manager. Rational Clear Quest is a change-request management tool that tracks and manages defects and change requests throughout the development process. With Clear Quest, you can manage every type of change activity associated with software development, including enhancement requests, defect reports, and documentation modifications. Rational Purify is a comprehensive C/C+ + run-time error checking tool that automatically pinpoints run-time errors and memory leaks in all components of an application, including third-party libraries, ensuring that code is reliable Rational Quantify is an advanced performance profiler that provides application performance analysis, enabling developers to quickly find, prioritize and eliminate performance bottlenecks within an application. Rational Pure Coverage is a customizable code coverage analysis tool that provides detailed application analysis and ensures that all code has been exercised, preventing untested code from reaching the end-user. Rational Suite Performance Studio is a sophisticated tool for automating performance tests on client/server systems. A client/server system includes client applications accessing a database or application server, and browsers accessing a Web server. Performance Studio includes Rational Robot and Rational Load Test. Use Robot to record client/server conversations and store them in scripts. Use Load Test to schedule and play back the scripts.

Performance Testing Process & Methodology 112 -

Proprietary & Confidential

Rational Robot. Facilitates functional and performance testing by automating record and playback of test scripts. Allows you to write, organize, and run tests, and to capture and analyze the results. Rational Test Factory. Automates testing by combining automatic test generation with source-code coverage analysis. Tests an entire application, including all GUI features and all lines of source code. During playback, Rational Load Test can emulate hundreds, even thousands, of users placing heavy loads and stress on your database and Web servers. Rational Test categorizes test information within a repository by project. You can use the Rational Administrator to create and manage projects.

The tools that are to discussed here are Rational Administrator Rational Robot Rational Test Manager

18.2 Rational Administrator

What is a Rational Project? A Rational project is a logical collection of databases and data stores that associates the data you use when working with Rational Suite. A Rational project is associated with one Rational Test data store, one RequisitePro database, one Clear Quest databases, and multiple Rose models and RequisitePro projects, and optionally places them under configuration management. Rational administrator is used to create and manage rational repositories, users and groups and manage security privileges. How to create a new project?

Performance Testing Process & Methodology 113 -

Proprietary & Confidential

Open the Rational administrator and go to File->New Project. In the above window opened enter the details like Project name and location. Click Next. In the corresponding window displayed, enter the Password if you want to protect the project with password, which is required to connect to, configure or delete the project.

Click Finish. In the configure project window displayed click the Create button. To manage the Requirements assets connect to Requisite Pro, to manage test assets create associated test data store and for defect management connect to Clear quest database.

Performance Testing Process & Methodology 114 -

Proprietary & Confidential

Once the Create button in the Configure project window is chosen, the below seen Create Test Data store window will be displayed. Accept the default path and click OK button.

Performance Testing Process & Methodology 115 -

Proprietary & Confidential

Once the below window is displayed it is confirmed that the Test datastore is successfully created and click OK to close the window.

Click OK in the configure project window and now your first Rational project is ready to play with.

Performance Testing Process & Methodology 116 -

Proprietary & Confidential

Rational Administrator will display your TestProject details as below:

18.3 Rational Robot

Rational Robot to develop three kinds of scripts: GUI scripts for functional testing and VU and VB scripts for performance testing. Robot can be used to: Perform full functional testing. Record and play back scripts that navigate through your application and test the state of objects through verification points. Perform full performance testing. Use Robot and TestManager together to record and play back scripts that help you determine whether a multi-client system is performing within user-defined standards under varying loads.
Proprietary & Confidential -

Performance Testing Process & Methodology 117 -

Create and edit scripts using the SQABasic, VB, and VU scripting environments. The Robot editor provides color-coded commands with keyword Help for powerful integrated programming during script development. Test applications developed with IDEs such as Visual Basic, Oracle Forms, PowerBuilder, HTML, and Java. Test objects even if they are not visible in the application's interface. Collect diagnostic information about an application during script playback. Robot is integrated with Rational Purify, Quantify, and PureCoverage. You can play back scripts under a diagnostic tool and see the results in the log.

The Object-Oriented Recording technology in Robot lets you generate scripts quickly by simply running and using the application-under-test. Robot uses Object-Oriented Recording to identify objects by their internal object names, not by screen coordinates. If objects change locations or their text changes, Robot still finds them on playback. The Object Testing technology in Robot lets you test any object in the application-under-test, including the object's properties and data. You can test standard Windows objects and IDEspecific objects, whether they are visible in the interface or hidden.

18.4 Robot login window

Once logged you will see the robot window. Go to File-> New->Script

Performance Testing Process & Methodology 118 -

Proprietary & Confidential

In the above screen displayed enter the name of the script say First Script by which the script is referred to from now on and any description (Not mandatory).The type of the script is GUI for functional testing and VU for performance testing.

18.5 Rational Robot main window-GUI script

Performance Testing Process & Methodology 119 -

Proprietary & Confidential

The GUI Script top pane) window displays GUI scripts that you are currently recording, editing, or debugging. It has two panes: Asset pane (left) Lists the names of all verification points and low-level scripts for this script. Script pane (right) Displays the script.

The Output window bottom pane) has two tabs: Build Displays compilation results for all scripts compiled in the last operation. Line numbers are enclosed in parentheses to indicate lines in the script with warnings and errors. Console Displays messages that you send with the SQAConsoleWrite command. Also displays certain system messages from Robot.

To display the Output window: Click View Output. How to record a play back script? To record a script just go to Record->Insert at cursor Then perform the navigation in the application to be tested and once recording is done stop the recording. Record-> Stop

18.6 Record and Playback options

Go to Tools-> GUI Record options the below window will be displayed.

Performance Testing Process & Methodology 120 -

Proprietary & Confidential

In this window we can set general options like identification of lists, menus ,recording think time in General tab: Web browser tab: Mention the browser type IE or Netscape Robot Window: During recording how the robot should be displayed and hotkeys details Object Recognition Order: the order in which the recording is to happen . For ex: Select a preference in the Object order preference list.

If you will be testing C++ applications, change the object order preference to C++ Recognition Order.


Playback options
Proprietary & Confidential -

Go to Tools-> Playback options to set the options needed while running the script.
Performance Testing Process & Methodology 121 -

This will help you to handle unexpected window during playback, error recovery, mention the time out period, to manage log and log data.

18.7 Verification points

A verification point is a point in a script that you create to confirm the state of an object across builds of the application-under-test. During recording, the verification point captures object information (based on the type of verification point) and stores it in a baseline data file. The information in this file becomes the baseline of the expected state of the object during subsequent builds. When you play back the script against a new build, Robot retrieves the information in the baseline file for each verification point and compares it to the state of the object in the new build. If the captured object does not match the baseline, Robot creates an actual data file. The information in this file shows the actual state of the object in the build. After playback, the results of each verification point appear in the log in Test Manager. If a verification point fails (the baseline and actual data do not match), you can select the verification point in the log and click View Verification Point to open the appropriate Comparator. The Comparator displays the baseline and actual files so that you can compare them.
Performance Testing Process & Methodology 122 Proprietary & Confidential -

A verification point is stored in the project and is always associated with a script. When you create a verification point, its name appears in the Asset (left) pane of the Script window. The verification point script command, which always begins with Result =, appears in the Script (right) pane. Because verification points are assets of a script, if you delete a script, Robot also deletes all of its associated verification points. You can easily copy verification points to other scripts if you want to reuse them.


List of Verification Points

The following table summarizes each Robot verification point. Type Alphanumeric Clipboard File Comparison File Existence Menu Description Captures and compares alphabetic or numeric values. Captures and compares alphanumeric data that has been copied to the Clipboard. Compares the contents of two files. Checks for the existence of a specified file Captures and compares the text, accelerator keys, and state of menus. Captures up to five levels of sub-menus. Checks whether a specified module is loaded into a specified context (process), or is loaded anywhere in memory. Captures and compares the data in objects. Captures and compares the properties of objects. Captures and compares a region of the screen (as a bitmap). Captures a baseline of a Web site and compares it to the Web site at another point in time. Checks the content of a Web site with every revision and ensures that changes have not resulted in defects. Checks that the specified window is displayed before continuing with the playback Captures and compares the client area of a window as a bitmap (the menu, title bar, and border are not captured).

Module Existence

Object Data Object Properties Region Image Web Site Compare Web Site Scan Window Existence Window Image

Performance Testing Process & Methodology 123 -

Proprietary & Confidential

18.8 About SQABasic Header Files

SQABasic header files let you declare custom procedures, constants, and variables that you want to use with multiple scripts or SQABasic library source files. SQABasic files are stored in the SQABas32 folder of the project, unless you specify another location. You can specify another location by clicking Tools General Options. Click the Preferences tab. Under SQABasic path, use the Browse button to find the location. Robot will check this location first. If the file is not there, it will look in the SQABas32 directory. You can use Robot to create and edit SQABasic header files. They can be accessed by all modules within the project. SQABasic header files have the extension .sbh.

18.9 Adding Declarations to the Global Header File

For your convenience, Robot provides a blank header file called Global.sbh. Global.sbh is a project-wide header file stored in SQABas32 in the project. You can add declarations to this global header file and/or create your own. To open Global.sbh: 1.Click File Open SQABasic File. 2.Set the file type to Header Files (*.sbh). 3. Select global.sbh, and then click Open.

18.10Inserting a Comment into a GUI Script:

During recording or editing, you can insert lines of comment text into a GUI script. Comments are helpful for documenting and editing scripts. Robot ignores comments at compile time. To insert a comment into a script during recording or editing. 1. If recording, click the Display GUI Insert Toolbar button on the GUI Record toolbar.

If editing, position the pointer in the script and click the Display GUI Insert Toolbar button on the Standard toolbar. 2. 3. 4. Click the Comment button on the GUI Insert toolbar. Type the comment (60 characters maximum). Click OK to continue recording or editing.

Robot inserts the comment into the script (in green by default) preceded by a single quotation mark. For example:

' This is a comment in the script To change lines of text into comments or to uncomment text: 1. Highlight the text.
Proprietary & Confidential -

Performance Testing Process & Methodology 124 -


Click Edit Comment Line or Edit Uncomment Line.

18.11About Data pools

A datapool is a test dataset. It supplies data values to the variables in a script during script playback. Datapools let you automatically pump test data to virtual testers under high-volume conditions that potentially involve hundreds of virtual testers performing thousands of transactions. Typically, you use a datapool so that: Each virtual tester that runs the script can send realistic data (which can include unique data) to the server. A single virtual tester that performs the same transaction multiple times can send realistic data to the server in each transaction.


Using Datapools with GUI Scripts

If you are providing one or more values to the client application during GUI recording, you might want a datapool to supply those values during playback. For example, you might be filling out a data entry form and providing values such as order number, part name, and so forth. If you plan to repeat the transaction multiple times during playback, you might want to provide a different set of values each time. A GUI script can access a datapool when it is played back in Robot. Also, when a GUI script is played back in a TestManager suite, the GUI script can access the same datapool as other scripts. There are differences in the way GUI scripts and sessions are set up for datapool access: You must add datapool commands to GUI scripts manually while editing the script in Robot. Robot adds datapool commands to VU scripts automatically. There is no DATAPOOL_CONFIG statement in a GUI script. The SQADatapoolOpen command defines the access method to use for the datapool. Although there are differences in setting up datapool access in GUI scripts and sessions, you define a datapool for either type of script using TestManager in exactly the same way.

18.12Debug menu
The Debug menu has the following commands:
Performance Testing Process & Methodology 125 Proprietary & Confidential -

Go Go Until Cursor Animate Pause Stop Set or Clear Breakpoints Clear All Breakpoints Step Over Step Into Step Out Note: The Debug menu commands are for use with GUI scripts only.

18.13Compiling the script

When you play back a GUI script or VU script, or when you debug a GUI script, Robot compiles the script if it has been modified since it last ran. You can also compile scripts and SQABasic library source files manually. . To compile Do this The active script or library source file Click File Compile. All scripts and library source files in Click File Compile All. Use this if, the current project for example, you have made changes to global definitions that may affect all of your SQABasic files During compilation, the Build tab in the Output window displays compilation results and error messages with line numbers for all compiled scripts and library source files. The compilation results can be viewed in the Build tab of the Output window.

Performance Testing Process & Methodology 126 -

Proprietary & Confidential

18.14Compilation errors

After the script is created and compiled and errors fixed it can be executed. The results need to be analyzed in the Test Manager.

Performance Testing Process & Methodology 127 -

Proprietary & Confidential

19 Rational Test Manager

Test Manager is the open and extensible framework that unites all of the tools, assets, and data both related to and produced by the testing effort. Under this single framework, all participants in the testing effort can define and refine the quality goals they are working toward. It is where the team defines the plan it will implement to meet those goals. And, most importantly, it provides the entire team with one place to go to determine the state of the system at any time. In Test Manager you can plan, design, implement, execute tests and evaluate results. With Test manager we can Create, manage, and run reports. The reporting tools help you track assets such as scripts, builds, and test documents, and track test coverage and progress. Create and manage builds, log folders, and logs. Create and manage data pools and data types When the script execution is started the following window will be displayed. The folder in which the log is to stored and the log name needs to be given in this window.

Performance Testing Process & Methodology 128 -

Proprietary & Confidential

19.1 Test Manager-Results screen

In the Results tab of the Test Manager, you could see the results stored. From Test Manager you can know start time of the script and

Performance Testing Process & Methodology 129 -

Proprietary & Confidential

Performance Testing Process & Methodology 130 -

Proprietary & Confidential

20 Supported environments 20.1 Operating system

WinNT4.0 with service pack 5 Win2000 WinXP(Rational 2002) Win98 Win95 with service pack1

20.2 Protocols
Oracle SQL server HTTP Sybase Tuxedo SAP People soft

20.3 Web browsers

IE4.0 or later Netscape navigator (limited support)

20.4 Markup languages

HTML and DHTML pages on IE4.0 or later.

20.5 Development environments

Visual basic 4.0 or above Visual C++ Java Oracle forms 4.5 Delphi Power builder 5.0 and above The basic product supports Visual basic, VC++ and basic web pages. To test other types of application, you have to download and run a free enabler program from Rationals website. For more details visit

Performance Testing Process & Methodology 131 -

Proprietary & Confidential

21 Performance Testing
The performance testing is a measure of the performance characteristics of an application. The main objective of a performance testing is to demonstrate that the system functions to specification with acceptable response times while processing the required transaction volumes in real-time production database. The objective of a performance test is to demonstrate that the system meets requirements for transaction throughput and response times simultaneously. The main deliverables from such a test, prior to execution, are automated test scripts and an infrastructure to be used to execute automated tests for extended periods.

21.1 What is Performance testing?

Performance testing of an application is basically the process of understanding how the web application and its operating environment respond at various user load levels. In general, we want to measure the latency, throughput, and utilization of the web site while simulating attempts by virtual users to simultaneously access the site. One of the main objectives of performance testing is to maintain a web site with low latency, high throughput, and low utilization.

21.2 Why Performance testing?

Performance problems are usually the result of contention for, or exhaustion of, some system resource. When a system resource is exhausted, the system is unable to scale to higher levels of performance. Maintaining optimum Web application performance is a top priority for application developers and administrators. Performance analysis is also carried for various purposes such as: During a design or redesign of a module or a part of the system, more than one alternative presents itself. In such cases, the evaluation of a design alternative is the prime mover for an analysis. Post-deployment realities create a need for the tuning the existing system. A systematic approach like performance analysis is essential to extract maximum benefit from an existing system. Identification of bottlenecks in a system is more of an effort at troubleshooting. This helps to replace and focus efforts at improving overall system response. As the user base grows, the cost of failure becomes increasingly unbearable. To increase confidence and to provide an advance warning of potential problems in case of load conditions, analysis must be done to forecast performance under load.

Typically to debug applications, developers would execute their applications using different execution streams (i.e., completely exercise the application) in an attempt to find errors. When looking for errors in the application, performance is a secondary issue to features; however, it is still an issue.
Performance Testing Process & Methodology 132 Proprietary & Confidential -

21.3 Performance Testing Objectives

The objective of a performance test is to demonstrate that the system meets requirements for transaction throughput and response times simultaneously. This infrastructure is an asset and an expensive one too, so it pays to make as much use of this infrastructure as possible. Fortunately, this infrastructure is a test bed, which can be re-used for other tests with broader objectives. A comprehensive test strategy would define a test infrastructure to enable all these objectives be met. The performance testing goals are: End-to-end transaction response time measurements. Measure Application Server components performance under various loads. Measure database components performance under various loads. Monitor system resources under various loads. Measure the network delay between the server and clients

21.4 Pre-Requisites for Performance Testing

We can identify five pre-requisites for a performance test. Not all of these need be in place prior to planning or preparing the test (although this might be helpful), but rather, the list defines what is required before a test can be executed. First and foremost thing is The design specification or a separate performance requirements document should : Defines specific performance goals for each feature instrumented. Bases performance goals on customer requirements. Defines specific customer scenarios. that is

Quantitative, relevant, measurable, realistic, achievable requirements As a foundation to all tests, performance requirements should be agreed prior to the test. This helps in determining whether or not the system meets the stated requirements. The following attributes will help to have a meaningful performance comparison. Quantitative - expressed in quantifiable terms such that when response times are measured, a sensible comparison can be derived. Relevant - a response time must be relevant to a business process. Measurable - a response time should be defined such that it can be measured using a tool or stopwatch and at reasonable cost. Realistic - response time requirements should be justifiable when compared with the durations of the activities within the business process the system supports.
Performance Testing Process & Methodology 133 Proprietary & Confidential -

Achievable - response times should take some account of the cost of achieving them.

Stable system A test team attempting to construct a performance test of a system whose software is of poor quality is unlikely to be successful. If the software crashes regularly, it will probably not withstand the relatively minor stress of repeated use. Testers will not be able to record scripts in the first instance, or may not be able to execute a test for a reasonable length of time before the software, middleware or operating systems crash. Realistic test environment The test environment should ideally be the production environment or a close simulation and be dedicated to the performance test team for the duration of the test. Often this is not possible. However, for the results of the test to be realistic, the test environment should be comparable to the actual production environment. Even with an environment which is somewhat different from the production environment, it should still be possible to interpret the results obtained using a model of the system to predict, with some confidence, the behavior of the target environment. A test environment which bears no similarity to the actual production environment may be useful for finding obscure errors in the code, but is, however, useless for a performance test.

21.5 Performance Requirements

Performance requirements normally comprise three components: Response time requirements Transaction volumes detailed in Load Profiles Database volumes

Response time requirements When asked to specify performance requirements, users normally focus attention on response times, and often wish to define requirements in terms of generic response times. A single response time requirement for all transactions might be simple to define from the users point of view, but is unreasonable. Some functions are critical and require short response times, but others are less critical and response time requirements can be less stringent. Load profiles The second component of performance requirements is a schedule of load profiles. A load profile is the level of system loading expected to occur during a specific business scenario. Business scenarios might cover different situations when the users organization has different levels of activity or involve a varying mix of activities, which must be supported by the system. Database volumes Data volumes, defining the numbers of table rows which should be present in the database after a specified period of live running complete the load profile. Typically, data volumes estimated to exist after one years use of the system are used, but two year volumes or greater might be used in some circumstances, depending on the business application.
Performance Testing Process & Methodology 134 Proprietary & Confidential -

22 Performance Testing Process

Requirements Collection Preparation Requirement Collection

Test Plan Preparation

Test Plan

Test Design Preparation

Test Design


Test Scripts

Test Execution

Pre Test & Post Test Procedure

Test Analysis

Preliminary Report

Is Performan ce Goal Reached?




Preparation of Reports Final Report

Performance Testing Process & Methodology 135 -

Proprietary & Confidential

22.1 Phase 1 Requirements Study

This activity is carried out during the business and technical requirements identification phase. The objective is to understand the performance test requirements, Hardware & Software components and Usage Model. It is important to understand as accurately and as objectively as possible the nature of load that must be generated. Following are the important performance test requirement that needs to be captured during this phase. Response Time Transactions Per Second Hits Per Second Workload No of con current users Volume of data Data growth rate Resource usage Hardware and Software configurations Activity Performance- Stress Test, Load Test, Volume Test, Spike Test, Endurance Test Work items Understand the system and application model Server side and Client side Hardware and software requirements. Browser Emulation and Automation Tool Selection Decide on the type and mode of testing Operational Inputs Time of Testing, Client and Server side parameters.

22.1.1 22.1.2 Deliverables


Deliverable Requirement Collection

RequirementCollectio n.doc

Performance Testing Process & Methodology 136 -

Proprietary & Confidential

22.2 Phase 2 Test Plan

The following configuration information will be identified as part of performance testing environment requirement identification. Hardware Platform Server Machines Processors Memory Disk Storage Load Machines configuration Network configuration Software Configuration Operating System Server Software Client Machine Software Applications Activity Test Plan Preparation Work items Hardware and Software Details Test data Transaction Traversal that is to be tested with sleep times. Periodic status update to the client.

Test Plan

Deliverable Sample


22.3 Phase 3 Test Design

Based on the test strategy detailed test scenarios would be prepared. During the test design period the following activities will be carried out: Scenario design Detailed test execution plan Dedicated test environment setup Script Recording/ Programming Script Customization (Delay, Checkpoints, Synchronizations points) Data Generation
Performance Testing Process & Methodology 137 Proprietary & Confidential -

Parameterization/ Data pooling

Activity Test Design Generation

Work items Hardware and Software requirements that includes the server components , the Load Generators used etc., Setting up the monitoring servers Setting up the data Preparing all the necessary folders for saving the results as the test is over. Pre Test and Post Test Procedures


Deliverable Sample

Test Design

22.4 Phase 4 Scripting

Activity Scripting Work items Browse through the application and record the transactions with the tool Parameterization, Error Checks and Validations Run the script for single user for checking the validity of scripts


Deliverable Test Scripts
Sample Script.doc


Performance Testing Process & Methodology 138 -

Proprietary & Confidential

22.5 Phase 5 Test Execution

The test execution will follow the various types of test as identified in the test plan. All the scenarios identified will be executed. Virtual user loads are simulated based on the usage pattern and load levels applied as stated in the performance test strategy. The following artifacts will be produced during test execution period: Test logs Test Result Activity Test Execution Work items Starting the Pre Test Procedure scripts which includes start scripts for server monitoring. Modification of automated scripts if necessary Test Result Analysis Report preparation for every cycle


Deliverable Test Execution
Time Sheet.doc


Run Logs.doc

22.6 Phase 6 Test Analysis

Activity Test Analysis Work items Analyzing the run results and preparation of preliminary report.


Deliverable Sample

Performance Testing Process & Methodology 139 -

Proprietary & Confidential

Test Analysis
Preliminary Report.doc

22.7 Phase 7 Preparation of Reports

The test logs and results generated are analyzed based on Performance under various load, Transaction/second, database throughput, Network throughput, Think time, Network delay, Resource usage, Transaction Distribution and Data handling. Manual and automated results analysis methods can be used for performance results analysis. The following performance test reports/ graphs can be generated as part of performance testing: Transaction Response time Transactions per Second Transaction Summary graph Transaction performance Summary graph Transaction Response graph Under load graph Virtual user Summary graph Error Statistics graph Hits per second graph Throughput graph Down load per second graph Based on the Performance report analysis, suggestions on improvement or tuning will be provided to the design team: Performance improvements to application software, middleware, database organization. Changes to server system parameters. Upgrades to client or server hardware, network capacity or routing. Activity Preparation of Reports Work items Preparation of final report.


Deliverable Final Report
Final Report.doc


Performance Testing Process & Methodology 140 -

Proprietary & Confidential

22.8 Common Mistakes in Performance Testing

No Goals
No general purpose model

Goals =>Techniques, Metrics, Workload Not trivial Biased Goals To show that OUR system is better than THEIRS Analysts = Jury Unsystematic Approach Analysis without Understanding the Problem Incorrect Performance Metrics Unrepresentative Workload Wrong Evaluation Technique Overlook Important Parameters Ignore Significant Factors Inappropriate Experimental Design Inappropriate Level of Detail No Analysis Erroneous Analysis No Sensitivity Analysis Ignoring Errors in Input Improper Treatment of Outliers Assuming No Change in the Future Ignoring Variability Too Complex Analysis Improper Presentation of Results Ignoring Social Aspects Omitting Assumptions and Limitations

22.9 Benchmarking Lessons

Ever build needs to be measured. We should run the automated performance test suite against every build and compare the results against previous results. Also, we should run the performance test suite under controlled conditions from build to build. This typically means measuring performance on "clean" test environments. Performance issues must be identified as soon as possible to prevent further degradation. Performance goals needs to be ensured. If we decide to make performance a goal and a measure of the quality criteria for release, the management team must decide to enforce the
Performance Testing Process & Methodology 141 Proprietary & Confidential -

goals. Establish incremental performance goals throughout the product development cycle. All the members in the team should agree that a performance issue is not just a bug; it is a software architectural problem. Performance testing of Web services and applications is paramount to ensuring an excellent customer experience on the Internet. The Web Capacity Analysis (WebCAT) tool provides Web server performance analysis; the tool can also assess Internet Server Application Programming Interface and application server provider (ISAPI/ASP) applications. Creating an automated test suite to measure performance is time-consuming and laborintensive. Therefore, it is important to define concrete performance goals. Without defined performance goals or requirements, testers must guess, without a clear purpose, at how to instrument tests to best measure various response times. The performance tests should not be used to find functionality-type bugs. Design the performance test suite to measure response times and not to identify bugs in the product. Design the build verification test (BVT) suite to ensure that no new bugs are injected into the build that would prevent the performance test suite from successfully completing. The performance tests should be modified consistently. Significant changes to the performance test suite skew or make obsolete all previous data. Therefore, keep the performance test suite fairly static throughout the product development cycle. If the design or requirements change and you must modify a test, perturb only one variable at a time for each build. Strive to achieve the majority of the performance goals early in the product development cycle because: Most performance issues require architectural change. Performance is known to degrade slightly during the stabilization phase of the development cycle.

Achieving performance goals early also helps to ensure that the ship date is met because a product rarely ships if it does not meet performance goals. You should reuse automated performance tests Automated performance tests can often be reused in many other automated test suites. For example, incorporate the performance test suite into the stress test suite to validate stress scenarios and to identify potential performance issues under different stress conditions. Tests are capturing secondary metrics when the instrumented tests have nothing to do with measuring clear and established performance goals. Although secondary metrics look good on wall charts and in reports, if the data is not going to be used in a meaningful way to make improvements in the engineering cycle, it is probably wasted data. En sure that you know what you are measuring and why.

Performance Testing Process & Methodology 142 -

Proprietary & Confidential

Testing for most applications will be automated. Tools used for testing would be the tool specified in the requirement specification. The tools used for performance testing are Loadrunner 6.5 and Webload 4.5x

Performance Testing Process & Methodology 143 -

Proprietary & Confidential

23 Tools
23.1 LoadRunner 6.5
LoadRunner is Mercury Interactives tool for testing the performance of client/server systems. LoadRunner enables you to test your system under controlled and peak load conditions. To generate load, LoadRunner runs thousands of Virtual Users that are distributed over a network. Using a minimum of hardware resources, these Virtual users provide consistent. Repeatable and measurable load to execute your client/server system just as real users would. LoadRunners in depth reports and graphs provide the information that you need to evaluate the performance of your client/server system.

23.2 WebLoad 4.5

Webload is a testing tool for testing the scalability, functionality and performance of Webbased applications both Internet and Intranet. It can measure the performance of your application under any load conditions. Use WebLoad to test how well your web site will perform under real-world conditions by combining performance, load and functional tests or by running them individually. Webload supports HTTP1.0 and 1.1, including cookies, proxies, SSL, TSL, client certificates, authentifications, persistent connections and chunked transfer coding. Webload generates load by creating virtual clients that emulate network traffic. You create test scripts (called agendas) using Java Scripts that instruct those virtual clients about what to do. When Webload runs the test, it gathers results at a per-client, per-transaction and perinstance level from the computers that are generating the load. Webload can also gather information servers performance monitor. You can watch the results as they occur- Webload displays them in graphs and tables in real-time and you can save and export the results when the test is finished.

Performance Testing Process & Methodology 144 -

Proprietary & Confidential

Performance Testing Tools - summary and comparison

This table lists several performance testing tools available on the market. For your convenience we compared them based on cost and OS required.

Tool Name




Load test tool emphasizing easeof-use. Supports all

23.3 Architecture Benchmarking

Hardware Benchmarking - Hardware benchmarking is performed to size the application with the planned Hardware platform. It is significantly different from capacity planning exercise in that it is done after development and before deployment Software Benchmarking - Defining the right placement and composition of software instances can help in vertical scalability of the system without addition of hardware resources. This is achieved through software benchmark test.

23.4 General Tests

What follows is a list of tests adaptable to assess the performance of most systems. The methodologies below are generic, allowing one to use a wide range of tools to conduct the assessments. Methodology Definitions Result: provide information about what the test will accomplish. Purpose: explains the value and focus of the test, along with some simple background information that might be helpful during testing. Constraints: details any constraints and values that should not be exceeded during testing. Time estimate: a rough estimate of the amount of time that the test may take to complete. Type of workload: in order to properly achieve the goals of the test, each test requires a certain type of workload. This methodology specification provides information on the appropriate script of pages or transactions for the user.
Performance Testing Process & Methodology 145 Proprietary & Confidential -

Methodology: a list of suggested steps to take in order to assess the system under test. What to look for: contains information on behaviors, issues and errors to pay attention to during and after the test.

Performance Testing Process & Methodology 146 -

Proprietary & Confidential

24 Performance Metrics
The Common Metrics selected /used during the performance testing is as below Response time Turnaround time = the time between the submission of a batch job and the completion of its output. Stretch Factor: The ratio of the response time with single user to that of concurrent users. Throughput: Rate (requests per unit of time) Examples: Jobs per second Requests per second Millions of Instructions Per Second (MIPS) Millions of Floating Point Operations Per Second (MFLOPS) Packets Per Second (PPS) Bits per second (bps) Transactions Per Second (TPS) Capacity: Nominal Capacity: Maximum achievable throughput under ideal workload conditions. E.g., bandwidth in bits per second. The response time at maximum throughput is too high. Usable capacity: Maximum throughput achievable without exceeding a pre-specified response-time limit Efficiency: Ratio usable capacity to nominal capacity. Or, the ratio of the performance of an n-processor system to that of a one-processor system is its efficiency. Utilization: The fraction of time the resource is busy servicing requests. Average Fraction used for memory. As tests are executed, metrics such as response times for transactions, HTTP requests per second, throughput etc., should be collected. It is also important to monitor and collect the statistics such as CPU utilization, memory, disk space and network usage on individual web, application and database servers and make sure those numbers recede as load decreases. Cognizant has built custom monitoring tools to collect the statistics. Third party monitoring tools are also used based on the requirement.

24.1 Client Side Statistics

Running Vusers Hits per Second Throughput HTTP Status Code HTTP responses per Second Pages downloaded per Second Transaction response time Page Component breakdown time Page Download time
Proprietary & Confidential -

Performance Testing Process & Methodology 147 -

Component size Analysis Error Statistics Errors per Second Total Successful/Failed Transactions

24.2 Server Side Statistics

System Resources - Processor Utilization, Memory and Disk Space Web Server ResourcesThreads, Cache Hit Ratio Application Server ResourcesHeap size, JDBC Connection Pool Database Server ResourcesWait Events, SQL Queries Transaction Profiling Code Block Analysis

24.3 Network Statistics

Bandwidth Utilization Network delay time Network Segment delay time

24.4 Conclusion
Performance testing is an independent discipline and involves all the phases as the mainstream testing lifecycle i.e strategy, plan, design, execution, analysis and reporting. Without the rigor described in this paper, executing performance testing does not yield anything more than finding more defects in the system. However, if executed systematically with appropriate planning, performance testing can unearth issues that otherwise cannot be done through mainstream testing. It is very typical of the project manager to be overtaken by time and resource pressures leading not enough budget being allocated for performance testing, the consequences of which could be disastrous to the final system. There is another flip side of the coin. However there is an important point to be noted here. Before testing the system for performance requirements, the system should have been architected and designed for meeting the required performance goals. If not, it may be too late in the software development cycle to correct serious performance issues. Web-enabled applications and infrastructures must be able to execute evolving business processes with speed and precision while sustaining high volumes of changing and unpredictable user audiences. Load testing gives the greatest line of defense against poor performance and accommodates complementary strategies for performance management and monitoring of a production environment. The discipline helps businesses succeed in leveraging Web technologies to their best advantage, enabling new business opportunity lowering transaction costs and strengthening profitability. Fortunately, robust and viable solutions exist to help fend off disasters that result from poor performance. Automated load testing tools and services are available to meet the critical need of measuring and optimizing complex and dynamic application and infrastructure performance. Once these solutions are properly adopted and utilized, leveraging an ongoing, lifecycle-focused approach, businesses
Performance Testing Process & Methodology 148 Proprietary & Confidential -

can begin to take charge and leverage information technology assets to their competitive advantage. By continuously testing and monitoring the performance of critical software applications, business can confidently and proactively execute strategic corporate initiatives for the benefit of shareholders and customers alike.

Performance Testing Process & Methodology 149 -

Proprietary & Confidential

25 Load Testing
Load Testing is creation of a simulated load on a real computer system by using virtual users who submit work as real users would do at real client workstations and thus testing the systems ability to support such workload. Testing of critical web applications during its development and before its deployment should include functional testing to confirm to the specifications, performance testing to check if it offers an acceptable response time and load testing to see what hardware or software configuration will be required to provide acceptable response time and handle the load that will created by the real users of the system

25.1 Why is load testing important ?

Load Testing increases the uptime for critical web applications by helping you spot the bottlenecks in the system under large user stress scenarios before they happen in a production environment

25.2 When should load testing be done?

Load testing should be done when the probable cost of the load test is likely less than the cost of a failed application deployment. Thus a load testing is accomplished by stressing the real application under simulated load provided by virtual users.

Performance Testing Process & Methodology 150 -

Proprietary & Confidential

26 Load Testing Process

26.1 System Analysis
This is the first step when the project decides on load testing for its system. Evaluation of the requirements and needs of a system, prior to load testing will provide more realistic test conditions. For this one should know all key performance goals and objectives like number of concurrent connections, hits per second etc., Another important analysis of the system would also include the appropriate strategy for testing applications. It can be load testing or stress testing or capacity testing. Load Testing is used to test the application against a requested number of users. The objective is to determine whether the site can sustain a requested number of users with acceptable response times. Stress testing is nothing but load testing over extended periods of time to validate an applications stability and reliability. Similarly capacity testing is used to determine the maximum number of concurrent users that an application can manage. Hence for businesses capacity testing would be the benchmark to say that the maximum loads of concurrent users the site can sustain before the system fails.

Finally it should also be taken into consideration of the test tool which supports load testing by determining its multithreading capabilities and the creation of number of virtual users with minimal resource consumption and maximal virtual user count.

26.2 User Scripts

Once the analysis of the system is done the next step would be the creation of user scripts. A script recorder can be used to capture all the business processes into test scripts and this more often referred as virtual users or virtual user scripts. A virtual user is nothing but an emulated real user who drives the real application as client. All the business process should be recorded end to end so that these transactions will assist in breakdown of all actions and the time it takes to measure the performance of business process.

26.3 Settings
Run time settings should be defined the way the scripts should be run in order to accurately emulate real users. Settings can configure the number of concurrent connections, test run time, follow HTTP redirects etc., System response times also can vary based on the connection speed. Hence throttling bandwidth can emulate dial up connections at varying modem speeds (28.8 Kbps or 56.6 Kbps or T1 (1.54M) etc.

Performance Testing Process & Methodology 151 -

Proprietary & Confidential

26.4 Performance Monitoring

Every component of the system needs monitoring :the clients, the network, the webs server, the application server, the database etc., This will result in instantly identifying the performance bottle necks during load testing. But if the tools support real time monitoring then testers would be able to view the application performance at any time during the test. Thus running the load test scenario and monitoring the performance would accelerate the test process thereby producing a more stable application

26.5 Analyzing Results

The last but most important step in load testing is collecting and processing the data to resolve performance bottlenecks. The reports generated can be anything ranging from Number of hits, number of test clients, requests per second, socket errors etc., Hence analyzing the results will isolate bottle necks and determine which changes are needed to improve the system performance. After these changes are made the tests must re run the load test scenarios to verify adjustments. Load Testing with WAST Web Application Stress is a tool to simulate large number of users with a relatively small number of client machines. Performance data on an web application can be gathered by stressing the website and measuring the maximum requests per second that the web server can handle. The next step is to determine which resource prevents the requests per second from going higher, such as CPU, memory, or backend dependencies.

26.6 Conclusion
Load testing is the measure of an entire Web application's ability to sustain a number of simultaneous users and transactions, while maintaining adequate response times. It is the only way to accurately test the end-to-end performance of a Web site prior to going live. Two common methods for implementing this load testing process are manual and automated testing. Manual testing would involve
Performance Testing Process & Methodology 152 -

Coordination of the operations of users

Proprietary & Confidential -

Measure response times Repeat tests in a consistent way Compare results

As load testing is iterative in nature, the performance problems must be identified so that system can be tuned and retested to check for bottlenecks. For this reason, manual testing is not a very practical option. Today, automated load testing is the preferred choice for load testing a Web application. The testing tools typically use three major components to execute a test: A console, which organizes, drives and manages the load Virtual users, performing a business process on a client application Load servers, which are used to run the virtual users

With automated load testing tools, tests can be easily rerun any number of times and the results can be reported automatically. In this way, automated testing tools provide a more cost-effective and efficient solution than their manual counterparts. Plus, they minimize the risk of human error during testing.

Performance Testing Process & Methodology 153 -

Proprietary & Confidential

27 Stress Testing
27.1 Introduction to Stress Testing
This testing is accomplished through reviews (product requirements, software functional requirements, software designs, code, test plans, etc.), unit testing, system testing (also known as functional testing), expert user testing (like beta testing but in-house), smoke tests, etc. All these testing activities are important and each plays an essential role in the overall effort but, none of these specifically look for problems like memory and resource management. Further, these testing activities do little to quantify the robustness of the application or determine what may happen under abnormal circumstances. We try to fill this gap in testing by using stress testing. Stress testing can imply many different types of testing depending upon the audience. Even in literature on software testing, stress testing is often confused with load testing and/or volume testing. For our purposes, we define stress testing as performing random operational sequences at larger than normal volumes, at faster than normal speeds and for longer than normal periods of time as a method to accelerate the rate of finding defects and verify the robustness of our product. Stress testing in its simplest form is any test that repeats a set of actions over and over with the purpose of breaking the product. The system is put through its paces to find where it may fail. As a first step, you can take a common set of actions for your system and keep repeating them in an attempt to break the system. Adding some randomization to these steps will help find more defects. How long can your application stay functioning doing this operation repeatedly? To help you reproduce your failures one of the most important things to remember to do is to log everything as you proceed. You need to know what exactly was happening when the system failed. Did the system lock up with 100 attempts or 100,000 attempts?[1] Note that there are many other types of testing which have not mentioned above, for example, risk based testing, random testing, security testing, etc. We have found, and it seems they agree, that it is best to review what needs to be tested, pick multiple testing types that will provide the best coverage for the product to be tested, and then master these testing types, rather than trying to implement every testing type. Some of the defects that we have been able to catch with stress testing that have not been found in any other way are memory leaks, deadlocks, software asserts, and configuration conflicts. For more details about these types of defects or how we were able to detect them, refer to the section Typical Defects Found by Stress Testing. Table 1 provides a summary of some of the strengths and weaknesses that we have found with stress testing.

Performance Testing Process & Methodology 154 -

Proprietary & Confidential

Table 1 Stress Testing Strengths and Weaknesses Strengths Weakness

Find defects that no other type of test would find Using randomization increase coverage Test the robustness of the application Not real world situation Defects are not always reproducible One sequence of operations may catch a problem right away, but use another sequence may never find the problem Does not test correctness of system response to user input

Helpful at finding memory leaks, deadlocks, software asserts, and configuration conflicts

27.2 Background to Automated Stress Testing

Stress testing can be done manually - which is often referred to as monkey testing. In this kind of stress testing, the tester would use the application aimlessly like a monkey - poking buttons, turning knobs, banging on the keyboard etc., in order to find defects. One of the problems with monkey testing is reproducibility. In this kind of testing, where the tester uses no guide or script and no log is recorded, its often impossible to repeat the steps executed before a problem occurred. Attempts have been made to use keyboard spyware, video recorders and the like to capture user interactions with varying (often poor) levels of success. Our applications are required to operate for long periods of time with no significant loss of performance or reliability. We have found that stress testing of a software application helps in accessing and increasing the robustness of our applications and it has become a required activity before every software release. Performing stress manually is not feasible and repeating the test for every software release is almost impossible, so this is a clear example of an area that benefits from automation, you get a return on your investment quickly, and it will provide you with more than just a mirror of your manual test suite. Previously, we had attempted to stress test our applications using manual techniques and have found that they were lacking in several respects. Some of the weaknesses of manual stress testing we found were: 1. Manual techniques cannot provide the kind of intense simulation of maximum user interaction over time. Humans can not keep the rate of interaction up high enough and long enough. 2. Manual testing does not provide the breadth of test coverage of the product features/commands that is needed. People tend to do the same things in the same way over and over so some configuration transitions do not get tested. 3. Manual testing generally does not allow for repeatability of command sequences, so reproducing failures is nearly impossible. 4. Manual testing does not perform automatic recording of discrete values with each command sequence for tracking memory utilization over time critical for detecting memory leaks.

Performance Testing Process & Methodology 155 -

Proprietary & Confidential

With automated stress testing, the stress test is performed under computer control. The stress test tool is implemented to determine the applications configuration, to execute all valid command sequences in a random order, and to perform data logging. Since the stress test is automated, it becomes easy to execute multiple stress tests simultaneously across more than one product at the same time. Depending on how the stress inputs are configured stress can do both positive and negative testing. Positive testing is when only valid parameters are provided to the device under test, whereas negative testing provides both valid and invalid parameters to the device as a way of trying to break the system under abnormal circumstances. For example, if a valid input is in seconds, positive testing would test 0 to 59 and negative testing would try 1 to 60, etc. Even though there are clearly advantages to automated stress testing, it still has its disadvantages. For example, we have found that each time the product application changes we most likely need to change the stress tool (or more commonly commands need to be added to/or deleted from the input command set). Also, if the input command set changes, then the output command sequence also changes given pseudo-randomization. Table 2 provides a summary of some of these advantages and disadvantages that we have found with automated stress testing.

Table 2 Automated Stress Testing Advantages and Disadvantages Advantages Disadvantages

Automated stress testing is performed under computer control Capability to test all product application command sequences Multiple product applications can be supported by one stress tool Uses randomization to increase coverage; tests vary with new seed values Repeatability of commands and parameters help reproduce problems or verify that existing problems have been resolved Informative log files facilitate investigation of problem Requires capital equipment and development of a stress test tool Requires maintaince of the tool as the product application changes Reproducible stress runs must use the same input command set Defects are not always reproducible even with the same seed value Requires test application information to be kept and maintained May take a long time to execute

In summary, automated stress testing overcomes the major disadvantages of manual stress testing and finds defects that no other testing types can find. Automated stress testing exercises various features of the system, at a rate exceeding that at which actual end-users can be expected to do, and for durations of time that exceed typical use. The automated stress test randomizes the order in which the product features are accessed. In this way, non-typical sequences of user interaction are tested with the system in an attempt to find latent defects not detectable with other techniques.
Performance Testing Process & Methodology 156 Proprietary & Confidential -

To take advantage of automated stress testing, our challenge then was to create an automated stress test tool that would: 1. Simulate user interaction for long periods of time (since it is computer controlled we can exercise the product more than a user can). 2. Provide as much randomization of command sequences to the product as possible to improve test coverage over the entire set of possible features/commands. 3. Continuously log the sequence of events so that issues can be reliably reproduced after a system failure. 4. Record the memory in use over time to allow memory management analysis. 5. Stress the resource and memory management features of the system.

27.3 Automated Stress Testing Implementation

Automated stress testing implementations will be different depending on the interface to the product application. The types of interfaces available to the product drive the design of the automated stress test tool. The interfaces fall into two main categories:


Programmable Interfaces: Interfaces like command prompts, RS-232, Ethernet, General Purpose Interface Bus (GPIB), Universal Serial Bus (USB), etc. that accept strings representing command functions without regard to context or the current state of the device.


Graphical User Interfaces (GUIs): Interfaces that use the Windows model to allow the user direct control over the device, individual windows and controls may or may not be visible and/or active depending on the state of the device.

27.4 Programmable Interfaces

These interfaces have allowed users to setup, control, and retrieve data in a variety of application areas like manufacturing, research and development, and service. To meet the needs of these customers, the products provide programmable interfaces, which generally support a large number of commands (1000+), and are required to operate for long periods of time, for example, on a manufacturing line where the product is used 24 hours a day, 7 days a week. Testing all possible combinations of commands on these products is practically impossible using manual testing methods. Programmable interface stress testing is performed by randomly selecting from a list of individual commands and then sending these commands to the device under test (DUT) through the interface. If a command has parameters, then the parameters are also enumerated by randomly generating a unique command parameter. By using a pseudorandom number generator, each unique seed value will create the same sequence of commands with the same parameters each time the stress test is executed. Each command is also written to a log file which can be then used later to reproduce any defects that were uncovered.
Performance Testing Process & Methodology 157 Proprietary & Confidential -

For additional complexity, other variations of the automated stress test can be performed. For example, the stress test can vary the rate at which commands are sent to the interface, the stress test can send the commands across multiple interfaces simultaneously, (if the product supports it), or the stress test can send multiple commands at the same time.

27.5 Graphical User Interfaces

In recent years, Graphical User Interfaces have become dominant and it became clear that we needed a means to test these user interfaces analogous to that which is used for programmable interfaces. However, since accessing the GUI is not as simple as sending streams of command line input to the product application, a new approach was needed. It is necessary to store not only the object recognition method for the control, but also information about its parent window and other information like its expected state, certain property values, etc. An example would be a HELP menu item. There may be multiple windows open with a HELP menu item, so it is not sufficient to simply store click the HELP menu item, but you have to store click the HELP menu item for the particular window. With this information it is possible to uniquely define all the possible product application operations (i.e. each control can be uniquely identified). Additionally, the flow of each operation can be important. Many controls are not visible until several levels of modal windows have been opened and/or closed, for example, a typical confirm file overwrite dialog box for a File->Save As filename operation is not available until the following sequence has been executed: 1. Set Context to the Main Window 2. Select File->Save As 3. Select Target Directory from tree control 4. Type a valid filename into the edit-box 5. Click the SAVE button 6. If the filename already exists, either confirm the file overwrite by clicking the OK button in the confirmation dialog or click the cancel button. In this case, you need to group these six operations together as one big operation in order to correctly exercise this particular OK button.

27.6 Data Flow Diagram

A stress test tool can have many different interactions and be implemented in many different ways. Figure 1 shows a block diagram, which can be used to illustrate some of the stress test tool interactions. The main interactions for the stress test tool include an input file and Device Under Test (DUT). The input file is used here to provide the stress test tool with a list of all the commands and interactions needed to test the DUT.

Performance Testing Process & Methodology 158 -

Proprietary & Confidential

System Resource Monitor

Input File

Stress Test Tool


Log command Sequence

Log Test Results

Figure 1: Stress Test Tool Interactions

Additionally, data logging (commands and test results) and system resource monitoring are very beneficial in helping determine what the DUT was trying to do before it crashed and how well it was able to manage its system resources. The basic flow control of an automated stress test tool is to setup the DUT into a known state and then to loop continuously selecting a new random interaction, trying to execute the interaction, and logging the results. This loop continues until a set number of interactions have occurred or the DUT crashes.

27.7 Techniques Used to Isolate Defects

Depending on the type of defect to be isolated, two different techniques are used: 1. System crashes (asserts and the like) do not try to run the full stress test from the beginning, unless it only takes a few minutes to produce the defect. Instead, back-up and run the stress test from the last seed (for us this is normally just the last 500 commands). If the defect still occurs, then continue to reduce the number of commands in the playback until the defect is isolated. 2. Diminishing resource issues (memory leaks and the like) are usually limited to a single subsystem. To isolate the subsystem, start removing subsystems from the database and re-run the stress test while monitoring the system resources. Continue this process until the subsystem causing the reduction in resources is identified. This technique is most effective after full integration of multiple subsystems (or, modules) has been achieved. Some defects are just hard to reproduce even with the same sequence of commands. These defects should still be logged into the defect tracking system. As the defect rePerformance Testing Process & Methodology 159 Proprietary & Confidential -

occurs, continue to add additional data to the defect description. Eventually, over time, you will be able to detect a pattern, isolate the root cause and resolve the defect. Some defects just seem to be un-reproducible, especially those that reside around page faults, but overall, we know that the robustness of our applications increases proportionally with the amount of time that the stress test will run uninterrupted.

Performance Testing Process & Methodology 160 -

Proprietary & Confidential

28 Test Case Coverage

28.1 Test Coverage

Test Coverage is an important measure of quality for software systems. Test Coverage analysis is the process of: Finding areas of a program not exercised by a set of test cases, Creating additional test cases to increase coverage, and Determining a quantitative measure of code coverage, which is an indirect measure of quality.

Also an optional aspect of test coverage analysis is: Identifying redundant test cases that do not increase coverage.

A test coverage analyzer automates this process. Test coverage analysis is sometimes called code coverage analysis. The two terms are synonymous. The academic world more often uses the term "test coverage" while practitioners more often use "code coverage". Test coverage analysis can be used to assure quality of the set of tests, and not the quality of the actual product. Coverage analysis requires access to test program source code and often requires recompiling it with a special command. Code coverage analysis is a structural testing technique (white box testing). Structural testing compares test program behavior against the apparent intention of the source code. This contrasts with functional testing (black-box testing), which compares test program behavior against a requirements specification. Structural testing examines how the program works, taking into account possible pitfalls in the structure and logic. Functional testing examines what the program accomplishes, without regard to how it works internally.

28.2 Test coverage measures

Performance Testing Process & Methodology 161 Proprietary & Confidential -

A large variety of coverage measures exist. Here is a description of some fundamental measures and their strengths and weaknesses
28.3 Procedure-Level Test Coverage

Probably the most basic form of test coverage is to measure what procedures were and were not executed during the test suite. This simple statistic is typically available from execution profiling tools, whose job is really to measure performance bottlenecks. If the execution time in some procedures is zero, you need to write new tests that hit those procedures. But this measure of test coverage is so coarse-grained it's not very practical.
28.4 Line-Level Test Coverage

The basic measure of a dedicated test coverage tool is tracking which lines of code are executed, and which are not. This result is often presented in a summary at the procedure, file, or project level giving a percentage of the code that was executed. A large project that achieved 90% code coverage might be considered a well-tested product. Typically the line coverage information is also presented at the source code level, allowing you to see exactly which lines of code were executed and which were not. This, of course, is often the key to writing more tests that will increase coverage: By studying the unexecuted code, you can see exactly what functionality has not been tested.
28.5 Condition Coverage and Other Measures

It's easy to find cases where line coverage doesn't really tell the whole story. For example, consider a block of code that is skipped under certain conditions (e.g., a statement in an if clause). If that code is shown as executed, you don't know whether you have tested the case when it is skipped. You need condition coverage to know. There are many other test coverage measures. However, most available code coverage tools do not provide much beyond basic line coverage. In theory, you should have more. But in practice, if you achieve 95+% line coverage and still have time and budget to commit to further testing improvements, it is an enviable commitment to quality!

28.6 How Test Coverage Tools Work

Performance Testing Process & Methodology 162 -

Proprietary & Confidential

To monitor execution, test coverage tools generally "instrument" the program by inserting "probes". How and when this instrumentation phase happens can vary greatly between different products. Adding probes to the program will make it bigger and slower. If the test suite is large and time-consuming, the performance factor may be significant.


Source-Level Instrumentation

Some products add probes at the source level. They analyze the source code as written, and add additional code (such as calls to a code coverage runtime) that will record where the program reached. Such a tool may not actually generate new source files with the additional code. Some products, for example, intercept the compiler after parsing but before code generation to insert the changes they need. One drawback of this technique is the need to modify the build process. A separate version namely, code coverage version in addition to other versions, such as debug (un optimized) and release (optimized) needs to be maintained. Proponents claim this technique can provide higher levels of code coverage measurement (condition coverage, etc.) than other forms of instrumentation. This type of instrumentation is dependent on programming language -- the provider of the tool must explicitly choose which languages to support. But it can be somewhat independent of operating environment (processor, OS, or virtual machine).


Executable Instrumentation

Probes can also be added to a completed executable file. The tool will analyze the existing executable, and then create a new, instrumented one. This type of instrumentation is independent of programming language. However, it is dependent on operating environment -- the provider of the tool must explicitly choose which processors or virtual machines to support.


Runtime Instrumentation

Probes need not be added until the program is actually run. The probes exist only in the in-memory copy of the executable file; the file itself is not modified. The same executable file used for product release testing should be used for code coverage. Because the file is not modified in any way, just executing it will not automatically start code coverage (as it would with the other methods of instrumentation). Instead, the code coverage tool must start program execution directly or indirectly. Alternatively, the code coverage tool will add a tiny bit of instrumentation to the executable. This new code will wake up and connect to a waiting coverage tool
Performance Testing Process & Methodology 163 Proprietary & Confidential -

whenever the program executes. This added code does not affect the size or performance of the executable, and does nothing if the coverage tool is not waiting. Like Executable Instrumentation, Runtime Instrumentation is independent of programming language but dependent on operating environment.

28.7 Test Coverage Tools at a Glance

There are lots of tools available for measuring Test coverage. Company Product OS Lang Win32, Bullseye BullseyeCoverage C/C++ Unix C/C++, CompuWare DevPartner Win32 Java, VB Win32, C/C++, Rational (IBM) PurifyPlus Unix Java, VB Software Win32, TCAT C/C++, Java Research Unix Win32, Testwell CTC++ C/C++ Unix Paterson LiveCoverage Win32 C/C++, VB Technology

Coverage analysis is a structural testing technique that helps eliminate gaps in a test suite. It helps most in the absence of a detailed, up-to-date requirements specification. Each project must choose a minimum percent coverage for release criteria based on available testing resources and the importance of preventing post-release failures. Clearly, safety-critical software should have a high goal. We must set a higher coverage goal for unit testing than for system testing since a failure in lower-level code may affect multiple high-level callers.

Performance Testing Process & Methodology 164 -

Proprietary & Confidential

29 Test Case points-TCP

29.1 What is a Test Case Point (TCP)

TCP is a measure of estimating the complexity of an application. This is also used as an estimation technique to calculate the size and effort of a testing project. The TCP counts are nothing but ranking the requirements and the test cases that are to be written for those requirements into simple, average and complex and quantifying the same into a measure of complexity. In this courseware we shall give an overview about Test Case Points and not elaborate on using TCP as an estimation technique.
29.2 Calculating the Test Case Points:

Based on the Functional Requirement Document (FRD), the application is classified into various modules like say for a web application, we can have Login and Authentication as a module and rank that particular module as Simple, Average and Complex based on the number and complexity of the requirements for that module. A Simple requirement is one, which can be given a value in the scale of 1 to3. An Average requirement is ranked between 4 and 7. A Complex requirement is ranked between 8 and 10. Complexity of Requirements Requirement Classification Simple (1-3) Average (4-7) Complex (> 8) Total

The test cases for a particular requirement are classified into Simple, Average and Complex based on the following four factors. Test case complexity for that requirement OR Interface with other Test cases OR No. of verification points OR Baseline Test data

Refer the test case classification table given below

Performance Testing Process & Methodology 165 Proprietary & Confidential - Test Case Classification Complexity Type Complexity of Test Case

Interface with other Test case 0 <3 >3

Number of verification points <2 3-8 >8

Baseline Test Data Not Required Required Required

Simple Average Complex

< 2 transactions 3-6 transactions > 6 transactions

A sample guideline for classification of test cases is given below. Any verification point containing a calculation is considered 'Complex' Any verification point, which interfaces with or interacts with another application is classified as 'Complex' Any verification point consisting of report verification is considered as 'Complex' A verification point comprising Search functionality may be classified as 'Complex' or 'Average' depending on the complexity

Depending on the respective project, the complexity needs to be identified in a similar manner. Based on the test case type an adjustment factor is assigned for simple, average and complex test cases. This adjustment factor has been calculated after a thorough study and analysis done on many testing projects. The Adjustment Factor in the table mentioned below is pre-determined and must not be changed for every project.

Performance Testing Process & Methodology 166 -

Proprietary & Confidential

Test Case Type Simple Average Complex Total Test Case Points

Complexity Adjustment Weight Factor 1 2 3 2(A) 4(B) 8(C)

Number Result No of Simple requirements in Number*Adjust factor A the project (R1) No of Average requirements in Number*Adjust factor B the project (R2) No of Complex requirements in Number*Adjust factor C the project (R3) R1+R2+R3

From the break up of Complexity of Requirements done in the first step, we can get the number of simple, average and complex test case types. By multiplying the number of requirements with it s corresponding adjustment factor, we get the simple, average and complex test case points. Summing up the three results, we arrive at the count of Total Test Case Points.

29.3 Chapter Summary

This chapter covered the basics on What is Test Coverage Test Coverage measures How does Test coverage tools work List of Test Coverage tools What is TCP and how to calculate the Test Case Points for an application

Performance Testing Process & Methodology 167 -

Proprietary & Confidential