+

Chapter 6 External Memory

William Stallings, Computer Organization and Architecture, 9th Edition
+ Objectives
 How are disks organized?
 How to insure data stored in disks?
 How to increase disk accessing?
 After studying this chapter, you should be able to:
 Understand the key properties of magnetic disks.
 Understand the performance issues involved in magnetic
disk access.
 Explain the concept of RAID and describe the various
levels.
 Compare and contrast hard disk drives and solid disk
drives.
 Describe in general terms the operation of flash memory.
 Understand the differences among the different optical
disk storage media.
 Present an overview of magnetic tape storage technology.
+
Contents

 6.1 Magnetic Disk

 6.2 Raid
 6.3 Solid State Drives
 6.4 Optical Memory
 6.5 Magnetic Tape
+
6.1- Magnetic Disk
 A disk is a circular platter constructed of nonmagnetic
material, called the substrate-chất nền, coated with a
magnetizable material
 Traditionally the substrate has been an aluminium or
aluminium alloy (hợp kim nhôm) material
(Wiki)
 Recently glass substrates have been introduced

 Benefits of the glass substrate:

 Improvement in the uniformity of the magnetic film
surface to increase disk reliability
 A significant reduction in overall surface defects to help
reduce read-write errors
 Ability to support lower fly heights (cho phép mỏng hơn)
 Better stiffness (cứng) to reduce disk dynamics
 Greater ability to withstand(anti) shock and damage
Magnetic Read Data are recorded on and later
retrieved from the disk via a
conducting coil named the head
and Write • In many systems there are two heads, a read
head and a write head

Mechanisms • During a read or write operation the head is

stationary while the platter rotates beneath it

Electric pulses are sent to the write

head and the resulting magnetic The write mechanism
patterns are recorded on the surface exploits the fact that
below, with different patterns for electricity flowing through a
positive and negative currents coil produces a magnetic
+ field

An electric current in the

wire induces a magnetic
field across the gap, which
in turn magnetizes a small
The write head itself is area of the recording
made of easily medium
magnetizable material and
is in the shape of a
rectangular doughnut with Reversing the direction of
a gap along one side and a the current reverses the
few turns of conducting direction of the
wire along the opposite magnetization on the
side recording medium
Inductive Write/Magnetoresistive
Read Head

Inductive Write: Ghi cảm ứng điện từ

Magneto-resistive Read: đọc từ điện
N: North, S: South
 Disk
Data
Layout
Disk Allocation Unit in Windows:
Cluster
Disk Layout Methods Diagram

Characteristics: Read by yourself

Winchester Disk Format
Seagate ST506

CRC- cyclic redundancy check – Data for error checking

Synch. Byte: Byte for identify the beginning of data
+ Table 6.1: Physical Characteristics
of Disk Systems
+ Characteristics
 Fixed-head disk  Removable disk
 One read-write head per  Can be removed and
track replaced with another disk
 Heads are mounted on a  Advantages:
fixed ridged arm that  Unlimited amounts of data are
extends across all tracks available with a limited
number of disk systems
 Movable-head disk  A disk may be moved from
 One read-write head one computer system to
another
 Head is mounted on an arm
 Floppy disks and ZIP
 The arm can be extended cartridge disks are
or retracted examples of removable
disks
 Non-removable disk
 Permanently mounted in the  Double sided disk
disk drive
 Magnetizable coating is
 The hard disk in a personal applied to both sides of the
computer is a non-
removable disk platter
+
Multiple
Platters
 Tracks

Cylinders

+
+ The head mechanism
provides a classification of
Disk Classification
disks into three types
 The head must generate or Winchester Heads
sense an electromagnetic  Used in sealed drive assemblies
field of sufficient magnitude that are almost free of
to write and read properly contaminants (chất ô nhiễm)
 The narrower the head, the  Designed to operate closer to the
closer it must be to the disk’s surface than conventional
platter surface to function rigid (rời) disk heads, thus
 A narrower head means allowing greater data density
narrower tracks and
therefore greater data  Is actually an aerodynamic foil (lá)
density that rests lightly on the platter’s
surface when the disk is
 The closer the head is to the motionless
disk the greater the risk of  The air pressure generated by a
error from impurities or spinning disk is enough to make
imperfections the foil rise above the surface
Typical Hard Disk Parameters
+
Timing of Disk I/O Transfer

The actual details of disk I/O operation depend on the

computer system, the operating
system, and the nature of the I/O channel and disk
controller hardware.
+ Disk Performance
Parameters
 When the disk drive is operating the disk is rotating
at constant speed

 Toread or write the head must be positioned at the

desired track and at the beginning of the desired
sector on the track
 Track selection involves moving the head in a movable-head
system or electronically selecting one head on a fixed-head
system
 Once the track is selected, the disk controller waits until the
appropriate sector rotates to line up with the head

 Seek time
 On a movable–head system, the time it takes to position the
head at the track
+ Disk Performance
Parameters
 Rotational delay (rotational latency)
 The time it takes for the beginning of the sector
to reach the head

 Access time
 The sum of the seek time and the rotational delay
 The time it takes to get into position to read or write

 Transfer time
 Once the head is in position, the read or write operation is
then performed as the sector moves under the head
 This is the data transfer portion of the operation
+
Exercise
 Một ổ cứng có tốc độ quay 5400RPM, seek time trung bình 8.5ms,
512 sectors/track. Hỏi thời gian trung bình để đọc 1 sector:

 Thời gian quay 1 vòng = 1*60*1000ms/5400 = 11.1 ms

 T1 = seek time trung bình = 8.5ms

 T2= average rotational delay = thời gian quay ½ vòng tròn

= 11.1 ms/2 = 5.5 ms

 T3= transfer time= thời gian quay 1/512 vòng tròn

= 11.1 ms/512 = 0.02 ms

 T= T1 + T2 + T3 = 14.02 ms

 Check the result above

 +  Consists of 7 levels

6.2- RAID  Levels do not imply a hierarchical

relationship but designate different
design architectures that share
three common characteristics:
Redundant Array of
Independent Disks 1) Set of physical disk drives
viewed by the operating system
as a single logical drive

Redundant Array of 2) Data are distributed across the

physical drives of an array in a
Inexpensive Disks scheme known as striping

3) Redundant disk capacity is used

Long life to store parity information,
Availability which guarantees data
Performance  Parallel accessing recoverability in case of a disk
failure
Reliability  Backup, checking
 Multi-disk design
Table 6.3 RAID Levels
RAID Levels
0, 1, 2
 RAID
Levels
3, 4, 5, 6
Data Mapping for a RAID Level 0 Array
+ RAID Level 0
 Addresses the issues of request patterns of the host systemand layout of the data
 Impact of redundancy does not interfere with analysis

RAID 0 for High Data Transfer

Capacity
RAID 0 for High I/O Request Rate
 For applications to
experience a high transfer  For an individual I/O request for a
rate two requirements must small amount of data the I/O time
be met: is dominated by the seek time and
rotational latency
1. A high transfer capacity must
exist along the entire path  A disk array can provide high I/O
between host memory and execution rates by balancing the
the individual disk drives I/O load across multiple disks

2. The application must make  If the strip size is relatively large

multiple waiting I/O requests can
I/O requests that drive the be handled in parallel, reducing
disk array efficiently the queuing time for each request
+ RAID Level 1
Characteristics
 Differs from RAID levels 2 through
6 in the way in which redundancy Positive Aspects
is achieved
 A read request can be serviced by
either of the two disks that contains
 Redundancy is achieved by the
the requested data
simple expedient of duplicating
all the data
 There is no “write penalty”
 Data striping is used but each
 Recovery from a failure is simple,
logical strip is mapped to two
when a drive fails the data can be
separate physical disks so that
accessed from the second drive
every disk in the array has a
mirror disk that contains the same  Provides real-time copy of all data
data
 Can achieve high I/O request rates if
 RAID 1 can also be implemented the bulk of the requests are reads
without data striping, although
this is less common  Principal disadvantage is the cost
+ RAID Level 2

Characteristics Performance
 Makes use of a parallel access  An error-correcting code is calculated
technique across corresponding bits on each data
disk and the bits of the code are stored
 In a parallel access array all
in the corresponding bit positions on
member disks participate in
multiple parity disks
the execution of every I/O
request  Typically a Hamming code is used,
which is able to correct single-bit
 Spindles of the individual
errors and detect double-bit errors
drives are synchronized so
that each disk head is in the  The number of redundant disks is
same position on each disk at proportional to the log of the number of
any given time data disks
 Data striping is used  Would only be an effective choice in an
 Strips are very small, often as environment in which many disk errors
small as a single byte or word occur
+ RAID Level 3
Performance
Redundancy
 In the event of a drive failure, the parity drive
is accessed and data is reconstructed from
 Requires only a single
the remaining devices
redundant disk, no matter
how large the disk array  Once the failed drive is replaced, the
missing data can be restored on the new
 Employs parallel access,
drive and operation resumed
with data distributed in
small strips  In the event of a disk failure, all of the data
are still available in what is referred to as
 Instead of an error
reduced mode
correcting code, a simple
parity bit is computed for  Return to full operation requires that the
the set of individual bits in failed disk be replaced and the entire
the same position on all of contents of the failed disk be regenerated on
the data disks the new disk
 Can achieve very high data  In a transaction-oriented environment
transfer rates performance suffers
+ RAID Level 4

Characteristics

 Makes use of an independent

access technique Performance

 In an independent access array,  Involves a write penalty

each member disk operates when an I/O write request of
independently so that separate small size is performed
I/O requests can be satisfied in
parallel  Each time a write occurs the
array management software
 Data striping is used must update the user data
 Strips are relatively large the corresponding parity
bits
 To calculate the new parity the
array management software must  Thus each strip write
read the old user strip and the old involves two reads and two
parity strip writes
+ RAID Level 5 RAID Level 6

Characteristics Characteristics

 Organized in a similar fashion  Two different parity calculations are

to RAID 4 carried out and stored in separate
blocks on different disks
 Difference is distribution of
the parity strips across all  Advantage is that it provides
disks extremely high data availability

 A typical allocation is a round-  Three disks would have to fail within

robin scheme the mean time to repair (MTTR)
interval to cause data to be lost
 The distribution of parity (usually expressed in hours)
strips across all drives avoids
the potential I/O bottleneck  Incurs (bears) a substantial write
found in RAID 4 penalty because each write affects
two parity blocks
Table 6.4: RAID Comparison (page 1 of 2)
 Table 6.4
RAID
Comparison
(page 2 of 2)
 6.3-Solid State Drive (SSD)
A memory device made
with solid state
Two distinctive types
components that can be Flash memory
used as a replacement to of flash memory:
a hard disk drive (HDD)
A type of semiconductor NOR
memory used in many •The basic unit of access is a bit
consumer electronic •Provides high-speed random access
•Used to store cell phone operating
products including smart system code and on Windows
phones, GPS devices, MP3 computers for the BIOS program that
The term solid players, digital cameras, runs at start-up
and USB devices
state refers to
electronic
circuitry built NAND
with Cost and performance
• The basic unit is 16 or 32 bits
• Reads and writes in small blocks
semiconductors has evolved to the point • Used in USB flash drives, memory
cards, and in SSDs
where it is feasible to use • Does not provide a random-access
to replace HDDs external address bus so the data
must be read on a block-wise basis
 Flash Memory Operation

(a) In transistors, a small voltage applied to the control gate can be

used to control the flow of a large current between the source
and the drain (ống dẫn).
(b) In a flash memory cell, a second gate(floating gate, insulated by
a thin oxide layer) is added to the transistor. Initially, the floating
gate does not interfere with the operation of the transistor .
(c) Applying a large voltage across the oxide layer causes electrons
to tunnel through it and become trapped on the floating gate,
where they remain even if the power is disconnected.
 SSD Compared to HDD
Table
SSDs have the following advantages over HDDs:
6.5
 High-performance input/output operations per second (IOPS)

 Durability/ Longer lifespan

 Lower power consumption

 Quieter and cooler running capabilities Comparisons

 Lower access times and latency rates

+
+
SSD
Organization
+ Practical Issues
There are two practical issues peculiar to SSDs
that are not faced by HDDs:
 Flash memory becomes
 SDD performance has a
unusable after a certain number
tendency to slow down as the
of writes
device is used
 Techniques for prolonging life:
 The entire block must be
read from the flash memory  Front-ending the flash with a
cache to delay and group write
and placed in a RAM buffer
operations
 Before the block can be
 Using wear-leveling algorithms
written back to flash memory, that evenly distribute writes
the entire block of flash across block of cells
memory must be erased  Bad-block management
 The entire block from the techniques
buffer is now written back to  Most flash devices estimate their
the flash memory own remaining lifetimes so
systems can anticipate failure
and take preemptive action
+ 6.4- Optical Memory
Compact Disk Read-Only Memory
 Audio CD and the CD-ROM share a similar technology
 The main difference is that CD-ROM players are more rugged and
have error correction devices to ensure that data are properly transferred

 Production:
 The disk is formed from a resin (nhựa nhân tạo) such as polycarbonate
 Digitally recorded information is imprinted as a series of microscopic pits
(hố) on the surface of the polycarbonate. This is done with a finely focused,
high intensity laser to create a master disk
 The master is used, in turn, to make a die to stamp out copies onto
polycarbonate
 The pitted surface is then coated with a highly reflective surface, usually
aluminum or gold
 This shiny surface is protected against dust and scratches by a top
coat of clear acrylic
 Finally a label can be silkscreened onto the acrylic
Table 6. 6: Optical Disk Products
+
CD Operation
+
CD-ROM Block Format
+
 CD-ROM is appropriate for the distribution of CD-ROM
large amounts of data to a large number of users

 Because the expense of the initial writing process

it is not appropriate for individualized
applications
 The CD-ROM has two advantages:
 The optical disk together with the information
stored on it can be mass replicated inexpensively

 The optical disk is removable, allowing the disk

itself to be used for archival storage

 The CD-ROM disadvantages:

 It is read-only and cannot be updated

 It has an access time much longer than that of a

magnetic disk drive
+ CD Recordable CD Rewritable
(CD-R) (CD-RW)
 Write-once read-many

 Accommodates applications in  Can be repeatedly written and overwritten

which only one or a small
 Phase change disk uses a material that has
number of copies of a set of
two significantly different reflectivities in two
data is needed
different phase states
 Disk is prepared in such a way  Amorphous (vô định hình)state: Molecules exhibit
that it can be subsequently a random orientation that reflects light poorly
written once with a laser beam
of modest-intensity  Crystalline state: Has a smooth surface that
reflects light well
 Medium includes a dye (thuốc
 A beam of laser light can change the material
nhuộm) layer which is used to
change reflectivity and is from one phase to the other
activated by a high-intensity  Disadvantage is that the material eventually
laser and permanently loses its desirable
properties
 Provides a permanent record
of large volumes of user data  Advantage is that it can be rewritten
+Digital
Versatile
Disk
(DVD)
Đĩa Đa
năng Số
High-Definition (HD) Optical Disks

CD: 700MB/single side

DVD: 4.7GB/single layer/single side
HD DVD: 15 GB/single layer/single side
Blu-ray: 25GB/single layer/single side
+
6.5- Magnetic Tape
 Tape systems use the same reading and recording techniques as
disk systems

 Medium is flexible polyester tape coated with magnetizable

material

 Coating may consist of particles of pure metal in special binders

or vapor-plated metal films

 Data on the tape are structured as a number of parallel tracks

running lengthwise

 Serial recording
 Data are laid out as a sequence of bits along each track

 Data are read and written in contiguous blocks called physical

records

 Blocks on the tape are separated by gaps referred to as inter-

record gaps
+
Magnetic
Tape
Features
+ Table 6.7: LTO Tape Drives
Linear tape-open (LTO)
+
Exercises
6.1 What are the advantages of using a glass
substrate for a magnetic disk?

6.2 How are data written onto a magnetic disk?

6.3 How are data read from a magnetic disk?

6.4 Explain the difference between a simple CAV

system and a multiple zoned recording system.

6.5 Define the terms track, cylinder, and sector.

6.6 What is the typical disk sector size?

+
Exercises

6.7 Define the terms seek time, rotational delay,

access time, and transfer time.

6.8 What common characteristics are shared by all

RAID levels?

6.9 Briefly define the seven RAID levels.

6.10 Explain the term striped data.

6.11 How is redundancy achieved in a RAID system?

6.12 In the context of RAID, what is the distinction

between parallel access and independent access?
+ Summary
External Memory

Chapter 6
 RAID
 Magnetic disk
 RAID level 0
 Magnetic read and write
mechanisms  RAID level 1
 Data organization and  RAID level 2
formatting
 RAID level 3
 Physical characteristics
 RAID level 4
 Disk performance parameters
 RAID level 5
 Solid state drives  RAID level 6
 Flash memory
 SSD compared to HDD  Optical memory
 SSD organization  Compact disk
 Practical issues
 Digital versatile disk

 Magnetic tape  High-definition optical disks