Disk perf' parameters and scheduling Disk Caches

1
- Disk perf. parameters and scheduling- Disk
Caches
2
Model of I/O organization
3
Disk Performance Parameters

• To read or write, the disk head must be
  positioned at the desired track, at the beginning
  of the desired sector
• Access time
• seek time (to position the head at the desired
  track) e.g. 10 ms
• rotational delay/latency (beginning of the sector
  to reach the head on average time for half
  rotation) e.g. 3 ms
• Data transfer occurs as the sector moves under
  the head
• transfer rate
• (bytes on a track)/(time for one rotation)
• e.g. 200Mbytes/sec

4
Disk Scheduling

• High emphasis disk-IO used in swapping, vmem
  key to the performance of the system
• Goals
• minimize overhead, i.e. seek time (seek
  distances)
• care for request response times

5
Disk Scheduling Policies FCFS

• Fair to all processes
• Approaches random scheduling in performance
• example 640 cylinder movements

6
Disk Scheduling PoliciesShorterst Seek Time
First

• Select the disk I/O request that requires the
  least movement of the disk arm from its current
  position
• may cause starvation of some requests
• Example 236 cylinder movements

7
Disk Scheduling PoliciesSCAN (aka elevator algo)

• Arm moves in one direction, satisfying all
  outstanding requests until it reaches the last
  track in that direction then direction is
  reversed
• example 236 cylinder movements
• Modification LOOK algo (dont go to the last
  track only that far where requests exist) would
  give 208 cylinder moves

8
Disk Scheduling PoliciesCircular SCAN

• Restricts scanning to one direction only
• When the last track has been visited in one
  direction, the arm is returned to the opposite
  end of the disk and the scan begins again
• Provides more uniform waiting time than SCAN

9
Disk Scheduling PoliciesAvoiding Arm Stickiness

• FSCAN
• Two queues
• One queue is empty for new requests while scan is
  performed on the other
• N-step-SCAN
• Segments the disk request queue into subqueues of
  length N
• Subqueues are serveded one at a time, using SCAN
• New requests added to other queue when queue is
  processed
• N1 gt FCFS N large gt SCAN

10
Selecting a Disk-Scheduling Algorithm

• SSTF is common SCAN, LOOK perform better for
  systems that place a heavy load on the disk.
• Performance depends on the number and types of
  requests.
• Requests for disk service can be influenced by
  the file-allocation method.
• Designer may introduce other priorities, e.g.
  swapping vs process IO, RT tasks, ...
• The disk-scheduler should be written as a
  separate module of the operating system, allowing
  it to be replaced with a different algorithm if
  necessary.

11
Disk Cache

• Buffer in main memory for disk sectors
• Contains a copy of some of the sectors on the
  disk
• LRU, LFU can be applied here without restrictions
  or special requirements (in contrast to the page
  replacement) why?

12
RAID Redundant Array of Independent Disks

• Replaces large capacity disk drives with multiple
  smaller capacity ones
• Distributes data so as to enable simultaneous
  access
• Goal
• improve IO performance
• allow easier incremental increases in capacity
• Why?
• Improvement in secondary storage performance has
  been considerably less than that in processors,
  main memory
• disk storage system can be a bottleneck

13
(No Transcript)
14
RAID 1 (mirrored)
15
RAID 2 (redundancy through Hamming code)
16
RAID 3 (bit-interleaved parity)
17
RAID 4 (block-level parity)
18
RAID 5 (block-level distributed parity)
19
RAID 6 (dual redundancy)