I/O buffering & disk scheduling

1. I/O Buffering
-Rushabh Shah

2. Buffering Techniques
 Single Buffer
 Double Buffer
 Circular Buffer

3. Single Buffer
 When a user process issues an I/O request, the operating system allocates a buffer
in the system portion of main memory to the operation.
 Input transfers are made to the system buffer. When the transfer is complete, the
process moves the block into user space and request another block.
 Thus this approach will generally speedup the I/O operations requested by a
process

5. Double Buffer
 An improvement over single buffering is by assigning two system buffers to the
operations.
 In this case, moving data from one system buffer to the process and reading data
from the I/O device to the other buffer may be performed simultaneously

7. Circular Buffer
 Double buffering is when more than two buffers are used and the collection of
these buffers is referred to as a circular buffer.
 It may be inadequate, if the process performs rapid burst of I/O. When two or more
buffers are used.
 In this way, although the I/O device is much slower, there may be enough data in
the system buffers for the process to read.

9. Disk Scheduling
-Rushabh Shah

10. Disk Scheduling
 A process needs two type of time, CPU time and IO time. For I/O, it requests the
Operating system to access the disk.
 However, the operating system must be fare enough to satisfy each request and at
the same time, operating system must maintain the efficiency and speed of process
execution.
 The technique that operating system uses to determine the request which is to be
satisfied next is called disk scheduling.

11. Purpose & Goal
Purpose:
 The main purpose of disk scheduling algorithm is to select a disk request from the
queue of IO requests and decide the schedule when this request will be processed.
Goal:
 Fairness
 High throughout
 Minimal traveling head time

12. Disk Scheduling Algorithms
 FCFS scheduling algorithm
 SSTF (shortest seek time first) algorithm
 SCAN scheduling
 C-SCAN scheduling
 LOOK Scheduling
 C-LOOK scheduling

13. FCFS Scheduling Algorithm
 It is the simplest Disk Scheduling algorithm. It services the IO requests in the order
in which they arrive. There is no starvation in this algorithm, every request is
serviced.
Disadvantages
 The scheme does not optimize the seek time.
 The request may come from different processes therefore there is the possibility of
inappropriate movement of the head.

14. Example
 Consider the following disk request sequence for a disk with 100 tracks 45, 21, 67,
90, 4, 50, 89, 52, 61, 87, 25
 Head pointer starting at 50 and moving in left direction. Find the number of head
movements in cylinders using FCFS scheduling.

15. Solution
Number of cylinders moved by the head
= (50-45)+(45-21)+(67-21)+
(90-67)+(90-4)+(50-4)+(89-50)+
(61-52)+(87-61)+(87-25)
= 5 + 24 + 46 + 23 + 86 + 46 +
49 + 9 + 26 + 62
= 376

16. SSTF Scheduling Algorithm
 Shortest seek time first (SSTF) algorithm selects the disk I/O request which requires
the least disk arm movement from its current position regardless of the direction. It
reduces the total seek time as compared to FCFS.
 It allows the head to move to the closest track in the service queue.
Disadvantages
 It may cause starvation for some requests.
 Switching direction on the frequent basis slows the working of algorithm.
 It is not the most optimal algorithm.

17. Example
 Consider the following disk request sequence for a disk with 100 tracks
 45, 21, 67, 90, 4, 89, 52, 61, 87, 25
 Head pointer starting at 50.
 Find the number of head movements in cylinders using SSTF scheduling.

18. Solution
• Number of cylinders = 5 + 7 + 9 + 6 + 20 + 2 + 1 + 65 + 4 + 17 =
136

19. I/O buffering
 The process of temporarily storing data that is passing between a processor and
a peripheral.
 The usual purpose is to smooth out the difference in rates at which the two
devices can handle data
 The zero-capacity case is sometimes referred to as a message system with no
buffering. The other cases are referred to as systems with automatic buffering.

20. Zero capacity.
 The queue has a maximum length of zero; thus, the link cannot have any messages waiting in it
Bounded capacity.
 The queue has finite length n; thus, at most n messages can reside in it. If the queue is not full
when a new message is sent, the message is placed in the queue.
Unbounded capacity.
 The queue’s length is potentially infinite; thus, any number of messages can wait in it. The sender
never blocks.