1 IO Management and Disk Scheduling Chapter 11

2 Categories of I/O Devices n Human readable u used to communicate with the user u video display terminals u keyboard u mouse u printer

3 Categories of I/O Devices n Machine readable u used to communicate with electronic equipment u disk drives u tape drives u controllers u Sensors u actuators u USB keys

4 Categories of I/O Devices n Communication u used to communicate with remote devices u digital line drivers u modems

5 Differences in I/O Devices n Data Transfer Rate: may be different of several orders of magnitude between the data transfer rates n Application u disk used to store files must have file- management software u disk used to store virtual memory pages needs special hardware to support it u terminal used by system administrator may have a higher priority

6 Differences in I/O Devices n Complexity of control n Unit of transfer u data may be transferred as a stream of bytes for a terminal or in larger blocks for a disk n Data representation u encoding schemes n Error conditions u devices respond to errors differently

7 I/O Device Data Rates

8 Techniques for Performing I/O n Programmed I/O u process is busy-waiting for the operation to complete n Interrupt-driven I/O u I/O command is issued u processor continues executing instructions u I/O module sends an interrupt when done

9 Techniques for Performing I/O n Direct Memory Access (DMA) u DMA module controls exchange of data between main memory and the I/O device u processor interrupted only after entire block has been transferred

10 Evolution of the I/O Function n Processor directly controls a peripheral device n Controller or I/O module is added u Processor uses programmed I/O without interrupts u Processor does not need to handle details of external devices

11 Evolution of the I/O Function n Controller or I/O module with interrupts u Processor does not spend time waiting for an I/O operation to be performed n Direct Memory Access u Blocks of data are moved into memory without involving the processor u Processor involved at beginning and end only

12 Evolution of the I/O Function n I/O module is a separate processor n I/O processor u I/O module has its own local memory u It is a computer in its own right

13 Direct Memory Address n Processor delegates I/O operation to the DMA module n DMA module transfers data directly to or from memory n When transfer is complete, DMA module sends an interrupt signal to the processor

14 DMA

15 DMA Configurations

16 DMA Configurations

17 Operating System Design Objectives – Non-functional Quality Attributes n Efficiency is an important issue u I/O cannot keep up with processor speed u Use of multiprogramming u Swapping is used to bring in additional Ready processes which is an I/O operation n Generality is an important issue u Desirable to handle all I/O devices in a uniform manner u Hide most of the details of device I/O in lower-level routines so that processes and upper levels see devices in general terms such as Read, Write, Open, and Close

18 A Model of I/O Organization Scheduling & Control Device I/O Scheduling & Control Scheduling & Control Local peripheral deviceCommunications portFile System Device I/O Device I/O Hardware User Processes User Processes User Processes User Processes User Processes User Processes Logical I/O Comm. Architecture Directory Management File System Physical Organization General access methods

19 I/O Buffering n Reasons for buffering u Processes must wait for I/O to complete before proceeding u Certain pages must remain in main memory during I/O

20 I/O Buffering n Block-oriented u information is stored in fixed sized blocks u transfers are made a block at a time u used for disks and tapes n Stream-oriented u transfer information as a stream of bytes u used for terminals, printers, communication ports, mouse, and most other devices that are not secondary storage

21 No Buffering Operating SystemUser Process No buffering I/O Device In

22 Single Buffer n Operating system assigns a buffer in main memory for an I/O request n Block-oriented u input transfers made to buffer u block moved to user space when needed u another block is moved into the buffer F read ahead Operating SystemUser Process Single buffering I/O Device InMove

23 Single Buffer n Block-oriented u user process can process one block of data while next block is read in u swapping can occur since input is taking place in system memory, not user memory u operating system keeps track of assignment of system buffers to user processes u output is accomplished by the user process writing a block to the buffer and later actually written out

24 Single Buffer n Stream-oriented u used a line at time u user input from a terminal is one line at a time with carriage return signaling the end of the line u output to the terminal is one line at a time

25 Double Buffer n Use two system buffers instead of one n A process can transfer data to or from one buffer while the operating system empties or fills the other buffer I/O Device In Move Operating SystemUser Process Double buffering

26 Circular Buffer n More than two buffers are used n Each individual buffer is one unit in a circular buffer n Used when I/O operation must keep up with process I/O Device InMove Operating SystemUser Process Circular buffering..

27 Disk Performance Parameters n To read or write, the disk head must be positioned at the desired track and at the beginning of the desired sector n Seek time u time it takes to position the head at the desired track n Rotational delay or rotational latency u time its takes until desired sector is rotated to line up with the head

28 Disk Performance Parameters n Access time u sum of seek time and rotational delay u the time it takes to get in position to read or write n Data transfer time u Data transfer occurs as the sector moves under the head u Data transfer for an entire file is faster when the file is stored in the same cylinder and in adjacent sectors

29 Disk Scheduling Policies n Seek time is the reason for differences in performance n Why disk scheduling? u For a single disk there will be a number of I/O requests u If requests are selected randomly, we will get the worst possible performance

30 Disk Scheduling Policies n First-in, first-out (FIFO) u process request sequentially u fair to all processes u approaches random scheduling in performance if there are many processes

31 Disk Scheduling Policies n Priority u goal is not to optimize disk use but to meet other objectives u short batch jobs may have higher priority u provide good interactive response time

32 Disk Scheduling Policies n Last-in, first-out u good for transaction processing systems F the device is given to the most recent user so there should be little arm movement u possibility of starvation since a job may never regain the head of the line

33 Disk Scheduling Policies n Shortest Service Time First u select the disk I/O request that requires the least movement of the disk arm from its current position u always choose the minimum Seek time

34 Disk Scheduling Policies n SCAN u arm moves in one direction only, satisfying all outstanding requests until it reaches the last track in that direction u direction is reversed

35 Disk Scheduling Policies n C-SCAN u restricts scanning to one direction only u 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

36 Disk Scheduling Policies n N-step-SCAN u segments the disk request queue into subqueues of length N u subqueues are process one at a time, using SCAN u new requests added to other queue when queue is processed n FSCAN u two queues u one queue is empty for new request

37 Disk Scheduling Algorithms

38 RAID – Redundant Array of Independent Disks n Logical view: Set of physical disk drives viewed by the operating system as a single logical drive n Physical view: Data are distributed across the physical drives of an array n Redundant disk capacity is used to store parity or redundant information n Why RAID? u Reliability u Performance

39 RAID 0 (non-redundant) strip 0 strip 4 strip 8 strip 12 strip 1 strip 5 strip 9 strip 13 strip 2 strip 6 strip 10 strip 14 strip 3 strip 7 strip 11 strip 15

40 Data Mapping for RAID Level 0 Array strip 0 strip 4 strip 8 strip 12 strip 1 strip 5 strip 9 strip 13 strip 2 strip 6 strip 10 strip 14 strip 3 strip 7 strip 11 strip 15 Physical Disk 0 Physical Disk 1 Physical Disk 2 Physical Disk 3 strip 0 strip 1 strip 2 strip 3 strip 4 strip 15 strip 14 strip 13 strip 12 strip11 strip 10 strip 9 strip 8 strip 7 strip 6 strip 5 Array Management Software A single logical disk drive A Set of physical disk drives

41 RAID 0 (nonredundant)

42 RAID 1 (mirrored)

43 RAID 2 (redundancy through Hamming code)

44 RAID 3 (bit-interleaved parity)

45 RAID 4 (block-level parity)

46 RAID 5 (block-level distributed parity)

47 RAID 6 (dual redundancy)

48 Disk Cache n Buffer in main memory for disk sectors n Contains a copy of some of the sectors on the disk

49 Least Recently Used n The block that has been in the cache the longest with no reference to it is replaced n The cache consists of a stack of blocks n Most recently referenced block is on the top of the stack n When a block is referenced or brought into the cache, it is placed on the top of the stack

50 Least Recently Used n The block on the bottom of the stack is removed when a new block is brought in n Blocks don’t actually move around in main memory n A stack of pointers is used

51 Least Frequently Used n The block that has experienced the fewest references is replaced n A counter is associated with each block n Counter is incremented each time block accessed

52 Least Frequently Used n Block with smallest count is selected for replacement n Some blocks may be referenced many times in a short period of time and the reference count is misleading

53 Disk Cache Performance

54 Disk Cache Performance