CHAPTER 6

EXTERNAL MEMORY
 GROUP 4
Noorfairuse bt mohamad noor 125718
Fatin najwa nursyadza 125693
m.Asyraf bin roslan 125709
Muhammad Firdaus Abdullah 125710
Muhammad Haris Azman bin Anuar 125712
Chia pei kiak 125690
Ng sing yee 125717
Chin wei chun 125692
Tan hui wen 125728
 MAGNETIC DISK

Magnetic READ AND WRITE MECHANISM

DATA ORGANIZATION AND FORMATTING

NOOR FAIRUSE BT MOHAMAD NOOR

125718
 MAGNETIC DISK
A disk is a circular platter constructed of nonmagnetic material, called
the substrate, coated with a magnetizable material.

Traditionally, the substrate has been an aluminium and aluminium alloy

material

More recently, glass substrate have been introduced

 MAGNETIC DISK
Benefits of glass substrate :
improvement in the uniformity of the magnetic film surface to increase disk
reliability.

A significant reduction in overall surface detects to help reduce read-write

errors.

Ability to support lower fly heights.

Better stiffness to reduce disk dynamics.

Greater ability to withstand shock and damage.

 MAGNETIC READ AND WRITE MECHANISM
Data are recorded on and later retrieved from the disk via a conducting coil named the
head.

In many systems there are two heads , a read head and write head

During a read or write operation the head is stationary while the platter rotates beneath
it
 The write head is made of easily magnetizable materials and is in the shape of a
rectangular doughnut shape with a gap along one side and a few turns of conducting
wire along the opposites side
 MAGNETIC READ AND WRITE
MECHANISM
Write Mechanism
The write mechanism exploits the fact that electricity flowing through a coil
produces a magnetic field

Electric pulse are send to the write head

The resulting magnetic patterns are recorded on the surface below, with
different patterns for positive and negative currents
 MAGNETIC READ AND WRITE
MECHANISM
Read Mechanism (Traditional)
Magnetic field moving relatives to a coil produces an electrical current in coil.

The structure of the head is the same as for writing mechanism.

Single head are used in floppy disk systems

MAGNETIC READ AND WRITE MECHANISM
Read Mechanism (Contemporary)
Use different read mechanism, requiring a separate read head, positioned for
convenience close to the write head.

Read head consist of a partially shielded magnetoresistive (MR) sensor.

Electrical resistance depends on the direction of the magnetization of the medium

moving under it.

Allows higher-frequency operation

greater storage densities
greater operating speed
 DATA ORGANIZATION AND FORMATTING
Concentric set of ring called tracks
there are gaps between adjacent tracks
same width as the head
prevent or minimizes errors due to misalignment of the head or inference of the
magnetic field
same number of bits per track (variable packing density)
Tracks divide in to sectors
Minimum block size is one sector
May have more than one sectors per block
Tracks divide into
sectors Concentric set
of rings
 DISK VELOCITY

PROBLEM :

Bit near centre of the rotating disk passes fixed point slower than bit on the outside of disk

SOLUTION :

Increasing between the spacing between bits of information recorded in the segment of
the disk
 Disk Layout Method Diagram : CAV
Constant angular velocity
Increase spacing betweens bits in different tracks
the information then can be scanned at the same rate by rotating the disk at a
fixed speed.
give pie shaped sectors and concentric tracks

Advantage Disadvantage

Individual tracks Waste of space on

and sectors outer tracks
addressable Lower data density
 Disk Layout Method Diagram : MZR
Multiple zoned recording
Surface is divided into a number of
concentric zones
Use zones to increase capacity
each zone has fixed bits per
track
more complex circuitry
 FINDING SECTORS : FORMATTING
Must be able to identify start track and sector

Format disk
formatted with some extra data used only by the disk drive and not accessible
to user
Winchester Disk Format
PHYSICAL CHARACTERISTICS
OF MAGNETIS DISK
BY FATIN NAJWA NURSYADZA BINTI MOHD RAMDAN
125693
DISK HEAD
FIXED-HEAD DISK
One read-write head per track (rare)
Heads mounted on a fixed rigid arm

MOVABLE-HEAD DISK
Only one read-write head per surface
Head mounted on movable arm
DISK
Nonremovable disk
Permanently mounted in disk drive
Example : Hard disk in personal computer
Removable disk
Can removed and replace from drive with another disk
Unlimited storage capacity
Easy data transfer between system
Example : Floppy disk and ZIP cartridge
SIDES
Double sided disks
The magnetizable coating is applied to both sides of
platter
Single sided disks
Less expensive disk system
PLATTERS
Multiple platters
Stack vertically
Movable head
One read-write head per platter surface
Heads are mechanically fixed and aligned
Aligned tracks on each platter form cylinder
Data is striped by cylinder
Reduce head movement
Increases speed (transfer rate)
HEAD MECHANISM
The head must generate or sense an electromagnetic field of
sufficient magnitude to write and read properly
The narrower the head , the closer it must be to the platter surface
to function
A narrower head means narrower tracks , therefore greater
data density
The closer the head to the disk the greater the risk of error from
impurities or imperfection
TYPE OF HEAD MECHANISMS

1.FIXED GAP
Traditionally , the read-write positioned at a fixed
distance above the platter, allowing an air gap
2. CONTACT (FLOPPY)
A head mechanism that actually comes into physical contact
with the medium during a read or write operation
8 ,5.25 , 3.5
Small capacity ( up to 1.44M byte )
Slow
Universal
Cheap
3. AERODYNAMIC GAP (WINCHESTER)

Used in sealed drive assemblies that are almost free of

contaminants
Designed to operate closer to the disks surface than conventional
rigid disk heads, thus allowing greater data density
Is actually an aerodynamic foil that rests lightly on the platters
surface when the disk is motionless
o The air pressure generated by a spinning disk is enough to make the foil
rise above the surface
SUMMARY
TYPE OF HEAD MOTION
1. FIXED HEAD ( ONE PER TRACK) TYPE OF PLATTERS
2. MOVABLE HEAD ( ONE PER SURFACE) 1. SINGLE PLATTER
TYPE OF DISK PORTABILITY 2. MULTIPLE PLATTER
1. NONREMOVABLE DISK TYPE OF HEAD MECHANISM
2. REMOVABLE DISK 1. CONTACT ( FLOPPY)
TYPE OF SIDES 2. FIXED GAP
1. SINGLE SIDED 3. AERODYNAMIC GAP ( WINCHESTER)
2. DOUBLE SIDED
Disk Performance
Parameters

M. Asyraf B Roslan
 Parameters
p

The disk I/O operation depends on

1. Computer system.
2. The operating system, and the nature of the I/O channel.
3. Disk controller hardware.

The disk operate at a rotating constant speed.

To read or write, the head must be positioned at the desired track

and at the beginning of the desired sector on that track.

Track selection involves moving the head in a movable-head

system.
The time it takes to position on tracks is called Seek Time.
Cont.

Once the track is selected, the disk controller waits until the
appropriate sector rotates to line up with the head.

The time it takes for the beginning of the sector to reach the head
is known as Rotational Delay

The sum of Seek Time and Rotational Delay will produce Access
Time
 Seek Time
Time required to move the disk arm to the required track.

Common disk size today is 3.5 inches for desktop and 2.5 inches for laptop.

Shrinking of disk diameter is to :

1. Improve seek performance

2. Enhanced rigidity
3. Mass reduction for faster spin
at stopped position
 Rotational Delay
The amount of time between information request and time
required to move to the sector located of the requested data.
It is a delay waiting for the rotation of the disk to bring to the
required sector .
Also called Rotational Latency.

20000/60 rps 1000/3 rps

1/(1000/3) 1 rev per 3ms
At 2000, the delay average 1.5ms
Example
Seek time: Say you're reading some data from the (0,0). You receive
instructions to read from track (2,5). The time it takes for you to move from
track 0 to track 2 is seek time.

Latency: Once you reach track 2, you realize the head is above the 1st
track and at sector 0. You'll have to wait till the disk rotates to the 5th
sector so that you can start reading from (2,5). The time you wait for the
sector to be accessible by your head here is known as latency.
TRANSFER TIME, T

The transfer time to / from the disk

depends on the rotation speed of the disk.
Following is the formula equation.
 There are 2 types of Accesses

Sequential access Random access

Sequential Access
2500 sector with 15000rpm
Seek time, 4ms per 500 sector Seek time, 4ms per 500 sector
15000rpm = 250rps
4ms per rev
2ms avg rot delay
Average seek = 4 ms
Average rot delay = 2 ms
Read 500 sectors = 4 ms
Total = 10 ms
Total = 10 + (4*6) = 34ms
Random Access
2500 sector with 15000rpm
15000rpm = 250rps
4ms per rev
2ms avg rot delay
Average seek = 4 ms
Average rot delay = 2 ms
Read 1 sector = 8 us
Total = 6.008 ms
Total = 2500 *6.008 = 15.02s
 R.A.I.D
</Redundant Array Independent Disk>

// Muhammad Firdaus Abdullah

 Rate of improvement:
secondary storage < processor & main memory
Bottlenecks processor

Use multiple parallel components (array of disks)

that operates independently + adding redundancy There are so many ways to
(to increase reliability of disk) organize the arrays
making it hard to develop
Single I/O request (distribute blocks of data across database schemes
multiple disks)
Multiple I/O request (data requested resides on
separate disks)
 :ENTER RAID:
Standardized industrial disk
for multiple- disk database
design
Levels (not hierarchical)
0-6 ( 7 levels)
levels = different organization
Redundant repetitive, too many
These levels has 3 similar aspect between them:
a) Physical arrayed disks are viewed as a single logical
drive by operating system.
b) Stripping Distribution of data across array of drives
c) Redundant disk capacity
- store parity info for data recovery in case of disk
failure
 RAID Strategy

Uses multiple disks & distributes data in a way

that data are accessed from multiple drives
simultaneously.

Need for redundancy rather than increasing number

of head and actuator (increase failure probability)
even though it can still improve I/O transfer rate.
 Does not include redundancy to improve performance
( For application that concerns performance rather than reliability)

High data transfer capacity

2 criteria:
-high transfer characteristics must exist along entire path
btwn host memory and disk drives.
-efficient I/O request application ( I/O request involves
parallel transfer of data from multiple disks increasing
effective transfer rate.

High I/O request rate

*response time are concerned rather than transfer rate.
*individual I/O (seek time + rotational latency).
*can provide high I/O execution rates by balancing I/O load across multiple disks.
 Logical disk is divided into strips in round
robin manner

Array management software maps logical to physical and

executed either in disk subsystem or host computer
 RAID Level 1
Have redundancy
-duplicating all the data (provides real - time copy of every data)
- each logical strip is mapped into TWO separate physical disks (mirror disks)
- costly (doubled disk number)
Positive aspect:
- read request can be serviced by whichever 2 disks that has
lower seek time + rotational latency
- no write penalty (unlike RAID 2 to 6 needs to compute then
update all parity bits and current strip)
- recovery from failure is simple

Performance can be increased further :

If application can split each read request so that both disk members can participate
 RAID Level 2
*Uses parallel access technique
*All members in disk array participate in every I/O request execution
*Data stripping are used as RAID 0 but strips are very small.
*Uses Hamming error correcting code to correct bit errors
(calculated across each data disks) and stored the corresponding bits
on multiple parity disks)
*Rather costly.
*Would only be effective if the environment has many occurring
disk errors.
 R.A.I.D
(LEVEL 3,4,5,6)

Muhammad Haris Azman bin Anuar

- It similar to RAID 2 , but only
requires only single redundant
disk
-data are striped in very
small strips
RAID 3
- Parallel transfer ,consists of byte-
level striping with a dedicated parity
disk

-drive failure, the parity drive is accessed and data is

reconstructed from the remaining devices

-the failed drive is replaced, the missing data

can be restored on the new drive and operation
resumed
-I/O invole the parallel transfer of
data
RAID 4 -consists of block-level striping with a
dedicated parity disk

-independent access technique , that mean

each disk operates indendently

-provides good performance of random reads, while

the performance of random writes is low due to the
need to write all parity data to a single disk

-every write operation must involve the parity disk,

which therefore can become a bottleneck
 -similar to RAID 4

-difference is that RAID 5 distributes the

parity strips across all disks
RAID 5
-unlike in RAID 4, parity information is
distributed among the drives.

-the distribution of parity strips across all drives , write

performance is increased and avoids the potential I/O
bottle-neck found in RAID 4
-RAID 6 extends RAID 5 by adding another parity block
RAID 6
-it uses block-level striping with two parity blocks
distributed across all member disks

-P and Q are two different data check algorithms (the exclusive-OR

calculation and independent data). This makes it possible to regenerate
data even if two disks containing user data fail

-have a performance penalty on write operations

because of the overhead associated with parity
calculations

-provides extremely high data availability

Level Advantages Disadvantages Applications

I/O performance is greatly improved by The failure of just one Video production and
spreading the I/O load across many channels drive will result in all editing
and drives data in an array Image Editing
0
No parity calculation overhead is involved being lost Pre-press applications
Very simple design Easy to implement Any application requiring
high bandwidth

100% redundancy of data means no rebuild is Highest disk overhead of Accountig Payroll
necessary in case of a disk failure, just a copy all RAID types (100%)
Financial
to the replacement disk inefficient
Any application requiring
1 Under certain circumstances, RAID 1 can
very high availability
sustain multiple simultaneous drive failures
Simplest RAID storage subsystem design
Level Advantages Disadvantages Applications

Extremely high data transfer rates possible
The higher the data transfer rate required,
the better the ratio of data disks to ECC
disks
2
Relatively simple controller design
compared to RAID levels 3, 4, & 5
Very high ratio of ECC disks No commercial
to data disks implementations exist/ not
with smaller word sizes commercially viable
inefficient
Entry level cost very high
requires very high transfer
rate requirement to justify

Very high read data transfer rate Very high Transaction rate equal to Video production and live
that of a single disk drive at streaming
write data transfer rate
best (if spindles are Image editing Video
Disk failure has an insignificant impact
3 synchronized)
on throughput editing
Controller design is fairly
Low ratio of ECC (parity) disks to data Prepress applications
complex
disks means high efficiency Any application requiring high
throughput
 Very high Read data transaction rate
Low ratio of ECC (parity) disks to data
disks means high efficiency
4 transfer rate
Highest Read data transaction rate
Low ratio of ECC (parity) disks to data
disks means high efficiency
5 Good aggregate transfer rate
Provides for an extremely high data
fault tolerance and can sustain mul-
6 tiple simultaneous drive failures
Quite complex controller No commercial
design implementations exist/ not
Worst write transaction commercially viable
rate and Write aggregate
Difficult and inefficient
data rebuild in the event
of disk failure
Most complex controller File and application servers
design
Database servers
Difficult to rebuild in the
Web, e-mail, and news
event of a disk failure (as
servers
compared to RAID level 1)
Intranet servers
Most versatile RAID level
More complex controller Perfect solution for mission
design critical applications
Controller overhead to
 Minimum number
Level Description
of drives[b]

RAID 0 Block-level striping without parity or mirroring 2

RAID 1 Mirroring without parity or striping 2

RAID 2 Bit-level striping with Hamming code for error correction 3

RAID 3 Byte-level striping with dedicated parity 3

RAID 4 Block-level striping with dedicated parity 3

RAID 5 Block-level striping with distributed parity 3

RAID 6 Block-level striping with double distributed parity 4

 -RAID 10 is the combination of RAID 0 and RAID
1. It applies parity check to realize striping set of
mirror. So it inherits speediness of RAID 0 and
security of RAID 1

-Higher reading and writing efficiency of RAID 0 and

higher data protection and recovery ability of RAID1
make RAID 10 a level with higher cost performance

-But the storage capacity utilization rate of

RAID 10 is as low as that of RAID, only 50%
SOLID STATE DRIVES(SSDs)

CHIA PEI KIAK

125690
Two Type of Solid State Drives

Internal SSDs External SSDs

Flash memory
Basic Operation of Flash Memory
 Basic Operation of Flash Memory Tunnel
Dielectric Oxide
In flash memory cell, a 2nd gate called a
floating gate, because it is insulated by a
thin layer oxide.
Initially, floating gate does not interfere
with the operation of transistor .
In this state, the cell is deemed to represent
binary 1
 Applying a large voltage(>5V) across the oxide
layer causes electron to tunnel through it and
become trapped on the floating gate, where they
remain even if the power is disconnect.
In this state, the cell is deemed to represent binary
0
The state of the cell can be read by external
circuitry to test whether the transistor is work or
not
Type of Flash Memory
i)NOR Flash Memory
Basic unit of access is a bit
Logical organization resembles a NOR logic device
Provides high-speed random access
Can read and write data to specific locations
Can reference and retrieve a single byte
Used to store cell phone operating system and on Windows
computer for the BIOS system
ii)NAND Flash Memory
Basic unit of access is 16 and 32 bits
Logical organization resembles a NAND logic device
Does not provides a random access external address bus
Can read and write data in small blocks
Provide higher bit density than NOR and greater write speed
Used in USB flash drives, memory cards
Flash memory
SSD Compare to HDD
SSDs have the following advantages over HDDs:
High-performance I/O operations per second(IOPS)
Durability
Longer lifespan
Lower power consumption
Quieter and cooler running capabilities
Lower access times and latency rates
SSD Organization
Host system is operating system invokes
file system software.
The file system will invokes I/O driver
software.
After invokes, software enable the host
system to interface with the SSD product.
If the SSD device is an internal hard
drive, a common interface used is PCIe.
If the SSD device is an external hard
drive, a common interface used is USB.
 Controller: Provides SSD device
level interfacing and firmware
execution.
Addressing: Logic that performs the
selection function across the flash
memory components.
Data buffer/cache: High speed RAM
memory components used for speed
matching and to increased data
throughput.
Error correction: Logic for error
detection and correction.
Flash memory components:
Individual NAND flash chips.
Practical Issues peculiar to SSDs
1.SSD performance has a tendency to slow down
The files are stored on disk as a set of pages.
Size page is 4KB in length
Flash memory is accessed in block. Block is 512KB which is
128 pages
Many file occupied in drive, the files become fragmented.

2.Unusable after a certain number of writes

When flash cells stressed, lose their ability to record and retain values
A typically limit is 100,000 writes
Presented by
Ng Sing Yee
 A computer memory that uses optical techniques.
Involves
Laser beam
Storage medium
A detector
 CD
Blue
Ray CD-
DVD ROM

DVD- Optical
RW Disk CD-R
Products

DVD-
R CD-
DVD RW
Compact Disk (CD)
Non erasable disk that stores digitized audio information.
Standard system uses 12-cm disks
Record > 60 min of uninterrupted playing time.

CD-ROM (Compact Disk Read-Only Memory)

Non erasable disk used for storing computer data.
Standard system uses 12-cm disks
Hold > 650 Mbytes.
 The main different between
Audio CD & CD-ROM

CD ROM are more rugged and have error correction devices to

ensure that the data are properly transferred from disk to computer.
- Data is stored in the form of land and pit.
- Laser shines through the clear polycarbonate while a motor spins the disk past it.
- Land smooth surface reflects high intensity light
- Pit rough surface reflects low intensity light
- Detected by photosensor and convert into digital signal.
- Beginning and end of pit = logic 1
- No change in elevation between interval = logic 0
To achieve greater capacity, CD and CD-ROM
Not organize information on concentric tracks (magnetic disk).
Disk contains a single spiral track, beginning near the center and spiraling
out the outer edge of the disk.
Information is packed evenly across the disk in segments in same size.
Pits are read by laser at a constant linear velocity(CLV).
Disk rotate more slowly for accesses near the outer edge than for those near
the center.
Capacity of track and the rotational delay both increase for position nearer
the outer edge of disk.
Data capacity for CD-ROM is about 680MB.
Sync : Identify beginning of a block
Header : Contains block address and mode byte
Mode 0 : blank data field
Mode 1 : error-correcting code(ECC) + 2048 bytes data
Mode 2 : 2336 bytes user data, no error correcting code
Data : User data
Auxiliary : Additional user data in mode 2
Advantages of CD-ROM:
Can be mass replicated inexpensively
The optical disk is removable, allowing the disk itself to be used
for archival storage

Disadvantage of CD-ROM:
Read only & cannot be updated
Longer access time than a magnetic disk
 Protective
acrylic

o Can write to the disk only once.

o Known as write-once read-many
o Permanent record of large volumes
of user data

How?
o Medium contain a dye layer (change the reflectivity)
o Activated by a high intensity laser
 Can erase and rewrite to the disk multiple times.

How?
Answer Use phase change technology /material
- A material that has two significantly different reflectivity
in two different phase states.
- Optical phase-change (a layer of metallic alloy)
Phase State
I. Crystalline - has a smooth surface
- reflects light well
- erased state
II. Amorphous- molecules in random orientation Protective
- reflects light poor acrylic
- recorded state

Strength of lasers
I. Highest: Write laser
II. Middle: Erase laser
III. Lowest: Read laser
Disadvantages of phase change
Have own erase cycles ( 500,000 and 1,000,000)
Burn and read speed is slower than a non-rewriteable disk.
Have shorter lives than magnetic platters under continuous use
DIGITAL
VERSATILE
DISK

Prepared by:
CHIN WEI CHUN
A LITTLE BIT ON DVD

Developed in year 1995

With 8 / 12cm diameters
Itis a replacement for some older
technologies
You should ask why!!!
WHAT DOES DVD REPLACE?

Analog VHS video tape

CD-ROM
ADVANTAGES OF DVD

DVD takes video into the digital age

It delivers movie with impressive picture quality
Can be randomly accessed like audio CDs,
which DVD machines can also play
Huge storage capacity and vivid quality
DVD VERSUS CD

DVD has greater capacity due to several

differences:
Bits are packed more closely on DVD
Double layers of pits on DVD
DVD-ROM can be two sided
DVD VERSUS CD
1. Bits are packed more closely on DVD
DVD uses a laser with shorter wavelength
Smaller loop spacing
(CD 1.6um, DVD 0.74um)
Smaller minimum distance between pits
(CD 0.834um, DVD 0.4um)
About a seven-fold increase in capacity
(CD 680MB, DVD 4.7GB)
DVD VERSUS CD
2. Double layers of pits on DVD
DVD employs a second layer of pits and land on
top of the first layer
A semireflective layer and a fully reflective layer
Lasers can read each layer separately by
adjusting focus.
Almost doubles the capacity of the disk
(4.7GB 8.5GB, not full doubling as second layer
has lower reflectivity)
DVD VERSUS CD
3. DVD-ROM can be two sided
One side with a fully reflective layer and a
semireflective layer too
Again duplicate the capacity (17GB)!!!!
However, only one side can be read at
once
The disk has to be flipped to read another
side
TYPES OF DVD
SUCCESSORS OF DVD
HIGH-DEFINATION OPTICAL DISKS
Why was it designed?
It is used to store high-definition videos and to provide
greater storage capacity compared to DVDs

Remember:
Greater capacity Higher bit density Shorter
wavelength laser blue-violet laser
HIGH-DEFINATION OPTICAL
DISKS
Types:
HD DVD (15GB on a single layer)
BLUE-RAY DVD (25GB on a single layer)

Having different disk formats

Brought to format wars
Blue-ray scheme achieved market dominance
BLUE-RAY DVD
Properties:
The data layer is positioned on the disk closer to
the laser
This enables a tighter focus and less distortion
Results in smaller pits and tracks.
Threeversions: read only (BD-ROM), recordable
once (BD-R), and rerecordable (BD-RE)
Smaller laser wavelength allows
a smaller minimum distance
between pits, hence greater bits
density on the disk
6.5 Magnetic Tape

Prepared by Tan Hui Wen

125728
Introduction

First kind of secondary memory.

Same reading and recording techniques as disk systems.
Dominant tape technology today is cartridge system
known as linear tape-open (LTO).
LTO was developed in late 1990s as an open-source
alternative to the various proprietary systems on the
market.
Parameters for various LTO generations
Usage of magnetic tape

Store large amount of data and long term storage and

archiving of data
As data backup to prevent data loss
Some batch-processing applications (eg. to hold the list of
data that will be processed)
Tape structure

Flexible polyester tape coated with particles of pure metal

Tape identical to the tape used in home recording
Tape width: 0.15 inch - 0.5 inch
Packaged as open reel that has to be threaded through
second spindle (reel)
Nowadays all tapes are housed in cartridges
Type of recording
Parallel recording
A byte of data may be recorded with one bit in each track
Earlier tape systems typically used 9 tracks, to store data
one byte at a time, 9th track is parity bit
Followed by tape systems using 18 or 36 tracks,
corresponding to a digital word or double word.
Serial recording
Data are laid out as sequence of bits along each track
Data are read and written in contiguous block on tape
(physical records)
Blocks on tape are separated by gaps (interrecord gaps)
Tape is formatted to assist in locating physical records.
Recording technique
Serpentine recording
When data are being recorded, first set of bits is recorded
on first track, until the end of the tape.
Next track is recorded after the heads are repositioned, in
opposite direction.
This procedure is repeated until all tracks are full.

Serpentine reading and writing

 The read-write head is able to read and write a number of
adjacent tracks simultaneously (typically 2-8 tracks)
Data are still recorded serially along each track, but blocks
(bits) in sequence are stored on adjacent tracks.

How to speed up
recording of
data?

Block layout for system that

reads-writes four tracks simultaneously.
Sequential-access device

To find a specific piece of data, the tape reader has to

start at the beginning of tape and continue forwarding
until it gets the desired data.
Eg. To read record N when tape head is positioned at
record 1, it has to read physical records 1 through N-1. If
the head is positioned beyond desired record, the tape
needs to be rewind a distance and restarted the reading
process.
Contrast to disk drives direct-access
Advantages
Small size
Provide low cost storage per GB
Large capacity up to TB

Disadvantages
Low speed write and read from (caused by serial access)