Silberschatz, Galvin and Gagne ©2013 Operating System Concepts Essentials – 2 nd Edition Chapter 6a: CPU Scheduling

Silberschatz, Galvin and Gagne ©2013Operating System Concepts Essentials – 2nd Edition

Chapter 6a: CPU Scheduling

6.2 Silberschatz, Galvin and Gagne ©2013Operating System Concepts Essentials – 2nd Edition

Chapter 6: CPU Scheduling

Basic Concepts

Scheduling Criteria

Scheduling Algorithms


Objectives

Introduce CPU scheduling Basis for multiprogrammed operating systems

Describe various CPU-scheduling algorithms

Discuss evaluation criteria for selecting a CPU-scheduling algorithm


Basic Concepts

Maximum CPU utilization with multiprogramming

CPU–I/O Burst Cycle Process execution = cycles of

CPU execution and I/O waits

CPU burst followed by I/O burst

CPU bursts are main concern


Histogram of CPU-burst Times


CPU Scheduler

Short-term scheduler selects among processes in ready queue, allocates CPU

Queue ordered in various ways Depends on algorithm


CPU Scheduler

Scheduling decisions may take place when process:

1. Switches from running to waiting state

2. Switches from running to ready state

3. Switches from waiting to ready

Terminates

1 and 4 are nonpreemptive Process voluntarily yields CPU

2 and 3 are preemptive Process preempted; CPU reallocated


Dispatcher

Dispatcher gives control of CPU to process selected by short-term scheduler; this involves: Switching context

Switching to user mode

Jumping to proper location in user program to restart

Dispatch latency – time dispatcher takes to stop one process and start another running



Basic Concepts

Scheduling Criteria



Scheduling Criteria

CPU utilization – keep CPU busy as possible

Throughput – # of processes that complete execution per time unit

Turnaround time – amount of time to execute a particular process


Scheduling Criteria

Waiting time –time process has been waiting in ready queue

Response time – amount of time from when request submitted until first response is produced Does not include device output


Scheduling Algorithm Optimization Criteria

Maximize CPU utilization

Maximize throughput

Minimize turnaround time

Minimize waiting time

Minimize response time



Basic Concepts

Scheduling Criteria



First- Come, First-Served (FCFS) Scheduling

Simplest algorithm

Process that requests CPU first is allocate CPU first

Easily manage by FIFO queue Process enters ready queue at tail CPU allocated to process at head Running process removed from queue Process runs to completion


First- Come, First-Served (FCFS) Scheduling

Process Burst Time

P1 24

P2 3

P3 3

Suppose processes arrive in order: P1 , P2 , P3

Gantt Chart for schedule:

Waiting time for P1 = 0; P2 = 24; P3 = 27 Average waiting time: (0 + 24 + 27)/3 = 17

P P P1 2 3

0 24 3027


FCFS Scheduling (Cont.)

Suppose processes arrive in order:

P2 , P3 , P1

Gantt chart schedule:

Waiting time for P1 = 6; P2 = 0; P3 = 3

Average waiting time: (6 + 0 + 3)/3 = 3

Much better than previous case

P 1

0 3 6 30

P 2 P 3


FCFS Scheduling (Cont.)

Convoy effect - short process behind long process E.g., one CPU-bound and many I/O-bound processes

All I/O bound processes wait in ready queue for CPU-bound to give up CPU

Lowers CPU and device utilization Compared to if shorter processes were allowed to go first


Shortest-Job-First (SJF) Scheduling

Each process knows length of next CPU burst CPU available => schedule process with smallest burst

Ties scheduled in FCFS

SJF is optimal – gives minimum average waiting time Difficulty: knowing length of next CPU request

Could ask user

Could make mathematical predictions E.g., .1 1 nnn t


Example of SJF

ProcessAril Time Burst Time

P1 0.0 6

P2 2.0 8

P3 4.0 7

P4 5.0 3

Average waiting time = (3 + 16 + 9 + 0) / 4 = 7

P 3

0 3 24

P 4 P 1

169

P 2


Example of Shortest-remaining-time-first

Add varying arrival times and preemption...

ProcessAarri Arrival TimeTBurst Time

P1 0 8

P2 1 4

P3 2 9

P4 3 5

Average waiting time = [(10-1)+(1-1)+(17-2)+5-3)]/4 = 26/4 = 6.5 msec

P 4

0 1 26

P 1 P 2

10

P 3P 1

5 17


Priority Scheduling

Priority number associated with each process SJF => priority scheduling algorithm

Priority = inverse of next CPU burst time

CPU allocated to process with highest priority E.g., smallest number = highest priority

Can be Preemptive or Nonpreemptive


Priority Scheduling

Priority Scheduling may result in starvation Low priority processes may never execute

Aging is possible solution Process’ priority increases with time


Example of Priority Scheduling

ProcessAarri Burst TimeT Priority

P1 10 3

P2 1 1

P3 2 4

P4 1 5

P5 5 2

Average waiting time = 8.2 msec


Round Robin (RR)

Each process gets unit of CPU time (time quantum q), usually 10-100 milliseconds After time elapsed, process is preempted, added to end

of ready queue

n processes in ready queue, time quantum q, Each process gets 1/n of CPU time

CPU time in chunks of at most q time units

No process waits more than (n-1)q time units


Round Robin (RR)

Timer interrupts every quantum to schedule next process If process finishes before time quantum, next process starts

If process not finished, added to tail of queue

Performance q large FIFO

q small q must be large with respect to context switch, otherwise overhead is too high


Example of RR with Time Quantum = 4

Time quantum = 4 ms

Process Burst Time

P1 24

P2 3

P3 3

Average wait time: 5.66 ms (6 + 4 + 7) / 3 = 17 / 3 = 5.66

P P P1 1 1

0 18 3026144 7 10 22

P 2 P 3 P 1 P 1 P 1


Example of RR with Time Quantum = 4

Typically, higher average turnaround than SJF But better response time

q should be large compared to context switch time q usually 10ms to 100ms, context switch < 10 µs


Time Quantum and Context Switch Time


Turnaround Time Varies With The Time Quantum

80% of CPU bursts should be shorter than q


Multilevel Queue

Ready queue partitioned into separate queues, e.g., foreground (interactive)

background (batch)

Processes cannot switch between queues Queue assignment is permanent

Each queue has own scheduling algorithm, e.g., foreground – RR

background – FCFS


Multilevel Queue: Scheduling b/t Qs

Fixed priority scheduling Serve all foreground then background

Possibility of starvation

Time slice Each queue gets certain amount of CPU time

Time scheduled amongst processes, e.g., 80% to foreground in RR

20% to background in FCFS


Multilevel Queue Scheduling


Multilevel Feedback Queue

Processes can move between various queues Aging implemented this way

Multilevel-feedback-queue scheduler defined by: Number of queues

Scheduling algorithms for each queue

Method to determine when to upgrade a process

Method to determine when to demote a process

Method to determine which queue a process will enter


Example of Multilevel Feedback Queue

Three queues Q0 – RR; q = 8 ms

Q1 – RR; 1 = 16 ms

Q2 – FCFS


Example of Multilevel Feedback Queue

New job enters Q0 , order is FCFS If process doesn’t finish in

8 ms, moved to Q1

Q1 order is FCFS

If job does not finish in 16 ms, moved to Q2

If no jobs in Q0 and Q1, jobs in Q2 processed FCFS

Documents

Silberschatz, Galvin and Gagne ©2013 Operating System Concepts Essentials – 2 nd Edition Chapter 6a: CPU Scheduling