|
welcome!
 Welcome to pecelleleosk.blogspot.my! We provide information and notes about Computer Organization and Architecture for you to access just by your fingertips! Easy and no stress! Enjoy your study :D recent entries
Credits
| Chapter 12:Memory Organization 2(External Memory)
Type of External
Memory
v Magnetic Disk
§ RAID
§ Removable
v Optical
§ CD-ROM
§ CD-Recordable (CD-R)
§ CD-R/W
§ DVD
v Magnetic Tape
Magnetic
Disk
Definition
A magnetic disk is a storage device that uses a magnetization
process to write, rewrite and access data. It is covered with a magnetic coating and
stores data in the form of tracks, spots and sectors. Hard disks, zip disks and floppy disks are common examples of magnetic disks.
v
Disk substrate coated with
magnetizable material (iron oxide…rust)
v
Substrate used to be
aluminium
v
Now glass
Ø Improved surface uniformity
v Increases reliability
Ø Reduction in surface defects
v Reduced read/write errors
Ø Lower flight heights (See later)
Ø Better stiffness
Ø Better shock/damage resistance
Read and Write Mechanisms
v
Recording
& retrieval via conductive coil called a head
v
May
be single read/write head or separate ones
v
During
read/write, head is stationary, platter rotates
v Write
Ø Current through coil produces
magnetic field
Ø Pulses sent to head
Ø Magnetic pattern recorded on
surface below
v Read (traditional)
Ø Magnetic field moving relative to
coil produces current
Ø Coil is the same for read and
write
v Read (contemporary)
Ø Separate read head, close to
write head
Ø Partially shielded magneto
resistive (MR) sensor
Ø Electrical resistance depends on
direction of magnetic field
Ø High frequency operation
v Higher storage density and speed
Data Organization and Formatting
v Concentric rings or tracks
Ø Gaps between tracks
Ø Reduce gap to increase capacity
Ø Same number of bits per track
(variable packing density)
Ø Constant angular velocity
v
Tracks
divided into sectors
v
Minimum
block size is one sector
v May have more than one sector per
block
Disk
Data Layout
Disk
Velocity
v
Bit
near centre of rotating disk passes fixed point slower than bit on outside of
disk
v
Increase
spacing between bits in different tracks
v Rotate disk at constant angular
velocity (CAV)
Ø Gives pie shaped sectors and
concentric tracks
Ø Individual tracks and sectors
addressable
Ø Move head to given track and wait
for given sector
Ø Waste of space on outer tracks
v Lower data density
v Can use zones to increase
capacity
Ø Each zone has fixed bits per
track
Ø More complex circuitry
Disk Layout Method Diagram
 
Tracks
and Cylinders
Memory
Access Methods
v Sequential
Ø Start at the beginning and read
through in order
Ø Access time depends on location
of data and previous location
Ø e.g. tape
v Direct
Ø Individual blocks have unique
address
Ø Access is by jumping to vicinity
plus sequential search
Ø Access time depends on location
and previous location
Ø e.g. disk
Media
Access Methods
v An
access method is a set of rules
governing how the network notice share the transmission medium. The rules for
sharing among computer are similar to the rules for sharing among the humans in
that they both boil down to a pair of fundamental philosophies:
Access
Method (2)
v Random
Ø Individual addresses identify
locations exactly
Ø Access time is independent of location
or previous access
Ø e.g. RAM
v Associative
Ø Data is located by a comparison
with contents of a portion of the store
Ø Access time is independent of
location or previous access
Ø e.g. cache
Memory Hierarchy
Definition
In computer architecture the memory hierarchy is
a concept used to discuss performance issues in computer architectural design,
algorithm predictions, and lower level programming constructs involving
locality of reference. The memory hierarchy in computer storage separates each of
its levels based on response time.
v Registers
§ In CPU
v Internal or Main memory
v External memory
Diagram
Performance
Access Time
v Time between presenting the address and getting the valid data
v Time interval between the instant at which an instruction control
unit initiates a call for data or a request to store data, and the instant at
which delivery of the data is completed or the storage is started.
Configuration
v Formatted Capacity, GB 250
v Sector Size, Byte 1024
v Data heads 10
v Data disks 5
v Performance
v Rotational speed, RPM 5400
v Disk transfer rate, MB/sec 100
v Controller overhead, μsec 30
v Seek time, ms 20
v Figure : Hard Disk Specification
Example:
Average
disk access time is the total times taken for average seek time + average
rotational delay + transfer time + controller overhead +queuing delay
=
20ms + (0.5/5400) + (1MB/100MB/s) + 30μs
=
20ms + 5.6ms + 10ms + 0.03ms = 35.63ms
However,
the manufacturer advertised average seek times are not the actual average seek
time.
Let
say the measured seek time is 50% of the advertised average seek time, the
average access time :
= 10ms +
5.6ms + 10ms + 0.03ms = 25.63ms
v However, the manufacturer advertised average seek times are not
the actual average seek time.
v Let say the measured seek time is 50% of the advertised average
seek time, the average access time :
= 10ms + 5.6ms + 10ms + 0.03ms = 25.63ms .
There
are 2 factors that mislead the manufacturer advertised seek time
v based on all possible seeks
v Locality and OS scheduling lead to smaller actual average seek
times
v Memory Cycle time
v Transfer Rate
RAID
v Redundant Array of Independent Disks
v Redundant Array of Inexpensive Disks
v 6 levels in common use
v Not a hierarchy
v Set of physical disks viewed as single logical drive by O/S
v Data distributed across physical drives
v Can use redundant capacity to store parity information
RAID
0 - Striping
v No redundancy
v Data striped across all disks
v Round Robin striping
v Increase speed
Advantages
Disadvantages
RAID
1 - Mirroring
v Mirrored Disks
v Data is striped across disks
v 2 copies of each stripe on separate disks
v Read from either
v Write to both
v Recovery is simple
v Expensive
Advantages
Disadvantages
RAID 2
v Disks are synchronized
v Very small stripes
v Error correction calculated across corresponding bits on disks
v Multiple parity disks store Hamming code error correction in
corresponding positions
v Lots of redundancy
v
This uses bit level striping. i.e
Instead of striping the blocks across the disks, it stripes the bits across the
disks.
v
In the above diagram b1, b2, b3 are
bits. E1, E2, E3 are error correction codes.
v
You need two groups of disks. One
group of disks are used to write the data, another group is used to write the
error correction codes.
v
This uses Hamming error correction
code (ECC), and stores this information in the redundancy disks.
v
When data is written to the disks, it
calculates the ECC code for the data on the fly, and stripes the data bits to
the data-disks, and writes the ECC code to the redundancy disks.
v
When data is read from the disks, it
also reads the corresponding ECC code from the redundancy disks, and checks
whether the data is consistent. If required, it makes appropriate corrections
on the fly.
v
This uses lot of disks and can be
configured in different disk configuration. Some valid configurations are 1) 10
disks for data and 4 disks for ECC 2) 4 disks for data and 3 disks for ECC
v
This is not used anymore. This is
expensive and implementing it in a RAID controller is complex, and ECC is
redundant now-a-days, as the hard disk themselves can do this.
Advantages
Disadvantages
RAID 3
Similar to RAID 2
v Only one redundant disk, no matter how large the array
v Simple parity bit for each set of corresponding bits
v Data on failed drive can be reconstructed from surviving data and
parity info
v Very high transfer rates
v
This uses byte level striping. i.e
Instead of striping the blocks across the disks, it stripes the bytes across
the disks.
v
In the above diagram B1, B2, B3 are
bytes. p1, p2, p3 are parities.
v
Uses multiple data disks, and a
dedicated disk to store parity.
v
The disks have to spin in sync to get
to the data.
v
Sequential read and write will have
good performance.
v
Random read and write will have worst
performance.
v
This is not commonly used.
Advantages
Disadvantages:
RAID
4
v Each disk operates independently
v Good for high I/O request rate
v Large stripes
v Bit by bit parity calculated across stripes on each disk
v Parity stored on parity disk
v
This uses block level striping.
v
In the above diagram B1, B2, B3 are
blocks. p1, p2, p3 are parities.
v
Uses multiple data disks, and a
dedicated disk to store parity.
v
Minimum of 3 disks (2 disks for data
and 1 for parity)
v
Good random reads, as the data blocks
are striped.
v
Bad random writes, as for every
write, it has to write to the single parity disk.
v
It is somewhat similar to RAID 3 and
5, but a little different.
v
This is just like RAID 3 in having
the dedicated parity disk, but this stripes blocks.
v
This is just like RAID 5 in striping
the blocks across the data disks, but this has only one parity disk.
v
This is not commonly used.
RAID 5
vTwo parity calculations
v Stored in separate blocks on different disks
v User requirement of N disks needs N+2
v High data availability
§ Three disks need to fail for data loss
§ Significant write penalty
Advantages
Disadvantages
RAID 6
v Two parity calculations
v Stored in separate blocks on different disks
v User requirement of N disks needs N+2
v High data availability
v
Just like RAID 5, this does block
level striping. However, it uses dual parity.
v
In the above diagram A, B, C are
blocks. p1, p2, p3 are parities.
v
This creates two parity blocks for
each data block.
v
Can handle two disk failure
v
This RAID configuration is complex to
implement in a RAID controller, as it has to
v
calculate two parity data for each
data block.
Advantages
Disadvantages
  |  |
0 Comments:
Post a Comment