A Presentation On

Design and Implementation of

Multiple Routing Configuration to
Detect Network Failure

Seminar by Guide

Sriharsha .M, Dr. K. Raghuveer,

USN: 5ZC08SCN22, Professor & Head,
M.Tech(CNE-FT), Dept of IS&E,
NIE – Mysore. NIE – Mysore.
 Outline
• Preface
• Existing System
• Proposed System
• Multiple Routing
• Network Recovery
• Requirements Specification
• Features of Multi Routing Configurations
• System Design
• System Implementation
• System Testing
• Conclusion and Future Scope
• Screenshots
 Preface
Problem Statement
Internet takes an increasingly central role in our communications
infrastructure, the slow convergence of routing protocols after a
network failure becomes a growing problem.

Causes of Network Failure

•Typical events that cause network failures are accidental cable
cuts, hardware malfunctions, software errors, natural disasters
(e.g., floods, fires, etc.), and human error (e.g., incorrect repair or
maintenance, etc.)
 Preface
• Failures occur frequently due to various causes such as faulty
interfaces, router crashes, routine maintenance, and accidental
fiber cuts, even in well-managed and well provisioned
networks.

• A study says 20% of the failures are due to planned

maintenance and more than 85% of the unplanned failures
affect only a single link or share a single router. Moreover, a
majority of these failures are transient: 46% last less than a
minute and 86% last less than ten minutes.
 Existing System
• In existing system IP networks are intrinsically robust, since
IGP routing protocols like OSPF are designed to update the
forwarding information based on the changed topology after a
failure.

• This re-convergence assumes full distribution of the new link

state to all routers in the network domain.

• When the new state information is distributed, each router

individually calculates new valid routing tables.
 Existing System
Limitations of Existing System
• Network-wide IP re-convergence is a time consuming process,
and a link or node failure is typically followed by a period of
routing instability.

• During this period, packets may be dropped due to invalid

routes and has an adverse effect on real-time applications.

• A key problem is that since most network failures are short

lived, too rapid triggering of the re-convergence process can
cause route flapping and increased network instability.
 Proposed System
• Proposed system exhibits recovery in all single failure
scenarios, using a single mechanism to handle both link and
node failures, and without knowing the root cause of the failure.

• System is strictly connectionless, and assumes only destination

based hop-by-hop forwarding.

• It is based on keeping additional routing information in the

routers, and allows packet forwarding to continue over an
alternative preconfigured output link immediately after the
detection of a failure.
 Proposed System
Advantages
• System is proactive and local protection mechanism, hence fast.

• It can be implemented with only minor changes to the existing

solutions.
 Multiple Routing
• More than one path is discovered for a given source-destination
pair.

• These multiple paths can be used either in case of failures as

backup paths to replace the primary path.

• They can be used simultaneously in the routing process of

different data packets via splitting.

• Multipath routing via splitting helps to increase the network

performance by balancing the network traffic and consequently
decreasing the queuing delay that the data packets experience.
 Multiple Routing
Multiple Routing Scenario
Area 0

SW0
1
2
3
F0/0

Area 2 R3 F1/0 Area 1

S1/0 S0/0
F0/0
FR1
S1/0

R4 3
1 FR0
S0/0
S0/1 2

S0/0
R5 S0/0 FR3
S0/2 S0/0

FR2

R6

External EIGRP Router

 Multiple Routing
Hop by Hop Forwarding
Destination Next Hop Destination Next Hop Destination Next Hop
10.1.0.0/24 R3 10.1.0.0/24 R1 10.1.0.0/24 R2
10.1.2.0/24 direct 10.1.2.0/24 R1 10.1.2.0/24 R2
10.2.1.0/24 direct 10.2.1.0/24 direct 10.2.1.0/24 R2
10.3.1.0/24 R3 10.3.1.0/24 R4 10.3.1.0/24 R2
20.2.0.0/16 R2 20.1.0.0/16 direct 20.1.0.0/16 R2
30.1.1.0/28 R2 20.2.1.0/28 direct 20.2.1.0/28 direct

H2

20.2.1.0/28 20.2.1.2/28
10.2.1.0/24 R2
R1

10.1.2.0/24 20.1.0.0/16

H1
10.3.0.0/16 R4
10.1.0.0/24 R3

Destination Next Hop Destination Next Hop Destination Next Hop

10.1.0.0/24 direct 10.1.0.0/24 direct 10.1.0.0/24 R3
10.1.2.0/24 R3 10.1.2.0/24 direct 10.1.2.0/24 R3
10.2.1.0/24 R3 10.2.1.0/24 R4 10.2.1.0/24 R2
10.3.1.0/24 R3 10.3.1.0/24 direct 10.3.1.0/24 direct
20.1.0.0/16 R3 20.1.0.0/16 R4 20.1.0.0/16 direct
20.2.1.0/28 R3 20.2.1.0/28 R4 20.2.1.0/28 R2
 Network Recovery
• Since network recovery is multi-dimensional in terms of
characteristics, many criteria are required to be considered for a
network recovery design. For example, the convergence time of
the recovery process, the cost of traffic rerouting, etc.

• In network recovery there are two main cycles- the recovery

cycle and, following it, the reversion cycle.

• The recovery cycle detects a fault and restores traffic onto the
backup paths. After it, the network is considered operational
again.

• After the fault repair, the reversion cycle redirects the traffic
from the backup path back to the working primary path.
 Network Recovery
Recovery and Reversion Process

• It is very common that backup paths are usually less optimal

than the primary path. However, considering the cost of
rerouting traffic, it may or may not be optimal to reroute all the
traffic back to the repaired primary path.
 Requirements Specification
Hardware Requirements
•PC (Intel Pentium Processor with more than 500 MHz)
•RAM - 256 MB
•Hard Disk - 40 GB or higher
•Monitor - Display panel (640 x 480)

Software Requirements
•SUN JDK 1.5 - Java development kit
•Eclipse Europa 3.3 – IDE
•Windows XP SP2 - Operating System
 Features of Multi Routing Configurations
• MRC is based on building a small set of backup routing
configurations that are used to route recovered traffic on
alternate paths after a failure.

MRC involves
• Creating backup configurations.

• Calculate configuration specific shortest paths and create

forwarding tables in each router.

• Forwarding process that takes advantage of the backup

configurations to provide fast recovery from a component
failure.
 Features of Multi Routing Configurations
• Construct configuration where link or node is not used to
forward traffic in a network.

• Packets that would normally be forwarded over the failed

interface are marked as belonging to a backup configuration,
and forwarded on an alternative interface towards its
destination.

• The packets must be marked with a configuration identifier, so

the routers along the path know which configuration to use.

• MRC does not affect the failure-free original routing.

• If a failure lasts for more than a specified time interval, a

normal re-convergence will be triggered.
 System Design
• System design is a process of problem solving and planning for
a software solution.

System Architecture
Client

Router A Router B Router C

Server
 System Design
System Flow Chart Source

Source sends file to destination

Routing is done

Multi routing configurations

No
Router
fails
Yes
Backup configuration is done

Connects to another router

Source side information is sent from

the damaged router

File transferred to destination

Destination
 System Design
• A use case diagram is a type of behavioral diagram defined by
UML created from a Use Case analysis.

Client Select file

View file
path

View file
size

Send file

View router
status
 System Design
Router
View file

View file
size

View
router
Status

Server
View files
received
 System Design
• Sequence Diagrams are an easy and intuitive way of describing
the behavior of a system by viewing the interaction between the
system and its environment.
Source File Processing Router Backup Destination
Configuration

File
Transmitting

Sending file
to router

Router
failure

Information
to source Transmitting
file using
backup
configuration
 System Design
Configuration Structure
•Configurations are defined by the network topology, which is
same in all configurations, and the associated link weights, which
differ among configurations.

•Network topology will be represented as graph G= (N, A), with

a set of nodes N and a set of unidirectional links (arcs) A.

•In the backup configurations, selected links and nodes must not
carry any transit traffic. Like “isolated” links that do not carry
traffic and “restricted” links that carry traffic only if no other
path available.
 System Design
Isolated Links and Nodes 2 3

1 4
•Isolated link: Infinite weight.
6 5

2 3
•Isolated node: All adjacent links at
least restricted. 1 4

Restricted link: High weight. 6 5

2 3
•Combinations possible.
Normal 1 4
Isolated
Restricted
23 6 5
 System Design
• Distinguish between the normal configuration as C0, and backup
configuration as Ci, i >0 can be made.

Normal Configuration 6
1

7
4
2

0 3

C0
 System Design
Building Configurations

6 6 6
1 1 1

7 7 7
4 4 4
2 2 2

5 5 5

0 3 0 3 0 3

C1 C2 C3
 System Design
Forwarding

6 6
1 1

7 7
4 4
2 2

5 5

0 3 0 3
 System Implementation
Modules
• Network module.
• Route configuration.
• Forwarding process.

Notations
G= (N, A) Graph comprising nodes N and directed links (arcs) A
Ci The graph with link weights as in configuration i
Si The set of isolated nodes in configuration Ci
Bi The backbone in configuration Ci
A (u) The set of links from node u
(u, v) The directed link from node u to v
pi (u, v) A given shortest path between nodes u and v in Ci
N (p) The nodes on path p
A (p) The links on path p
wi (u, v) The weight of link (u, v) in configuration Ci
wi (p) The total weight of the links in path p in configuration Ci
wr The weight of a restricted link
n The number of configurations to generate (algorithm input)
 System Implementation
Algorithm for Configuration
 System Implementation
Packet Forwarding
Packet arrives Switched
Yes
configuration Drop packet
before?
Normal lookup No
Lookup in neighbor’s
backup configuration
Output link Yes
failed?
Failed link Yes Lookup in own
No returned? backup configuration

No

Forward
29
 System Testing
Levels of Testing
•Unit Testing.
•Integration Testing.
•System and Acceptance Testing.

Test Cases-Unit Testing

Actual Output
Sl. No Test Case Input Expected Result Result
1 Client Server Socket connection Client request to Same as Pass
Communication established between client be processed excepted
and server
 System Testing
Blackbox Testing
Test Case Input Expected Output Actual Output Result
Sl. No
1 Send data from client1 to server Browse any file and click on send First identified router should fails to Same as expected Pass
via router process the request and acknowledge
the same
The next identified router will process Same as expected Pass
the request and acknowledge the same,
should display data
The server should receive the data Same as expected Pass

Send data from client2 to server Browse any file and click on send First identified router should fails to Same as expected Pass
2 via router process the request and acknowledge
the same
The next identified router will process Same as expected Pass
the request and acknowledge the same,
should display data
The server should receive the data Same as expected Pass

3 Send data simultaneously from
client1 and clent2 to server via
router
Browse any file from client1 and client2
click on send simultaneously
Request from client1 should be Same as expected Pass
processed and acknowledged by router.
Server is expected to receive data from
client1 first
As soon as server receives data from Same as expected Pass
client1. The request from client2 should
be processed by idle router. The server
is expected to receive data and display
the same form client2.
 Conclusion and Future Scope
• Assured single-fault tolerance.

• Covers link and node failures.

• Need not know the root cause of failure.

• Rapid, local recovery.

• Modest state overhead.

• Future scope- Load balancing, Multi-fault tolerance, Multicast

node-fault tolerance, Incremental topology changes, Loop-free
convergence.
Screenshots
Screenshots
Screenshots
Thank You…