Chapter 6 CPU Scheduling Operating System Concepts 9

Chapter 6: CPU Scheduling n Basic Concepts n Scheduling Criteria n Scheduling Algorithms n

Objectives n To introduce CPU scheduling, which is the basis for multiprogrammed operating systems

Basic Concepts n Maximum CPU utilization obtained with multiprogramming n CPU–I/O Burst Cycle –

Histogram of CPU-burst Times n Most often, we only look at one CPU-burst of

CPU Scheduler n Short-term scheduler selects from among the processes in ready queue, and

Dispatcher n Dispatcher module gives control of the CPU to the process selected by

Scheduling Criteria n CPU utilization – what percentage of CPU time is used for

Scheduling Algorithm Optimization Criteria n Max CPU utilization n Max throughput n Min turnaround

Chapter 6: CPU Scheduling n Basic Concepts n Scheduling Criteria n Scheduling Algorithms l

First- Come, First-Served (FCFS) Scheduling Process Burst Time P 1 24 P 2 3

FCFS Scheduling (Cont. ) Suppose that the processes arrive in the order at time

Shortest-Job-First (SJF) Scheduling n Associate with each process the length of its next CPU

Example of SJF Process. Arrival Time Burst Time P 1 0. 0 6 P

Determining Length of Next CPU Burst n Can only estimate the length – should

Prediction of the Length of the Next CPU Burst n CPU burst (ti): 6

Examples of Exponential Averaging n =0 n+1 = n l Recent history does not

Example of Shortest-remaining-time-first n Now we add the concepts of varying arrival times and

Priority Scheduling n A priority number (integer) is associated with each process n The

Example of Priority Scheduling Process. A arri Burst Time. T Priority P 1 10

Round Robin (RR) n Each process gets a small unit of CPU time (time

Example of RR with Time Quantum = 4 Process Burst Time P 1 P

Time Quantum and Context Switch Time Operating System Concepts – 9 th Edition 6.

Turnaround Time Varies With The Time Quantum n Four processes arrive at time 0

Multilevel Queue n Ready queue is partitioned into separate queues, e. g. , l

Multilevel Queue Scheduling Operating System Concepts – 9 th Edition 6. 28 Silberschatz, Galvin

Multilevel Feedback Queue n A process can move between the various queues; aging can

Example of Multilevel Feedback Queue n Three queues: l Q 0 – RR with

Multiple-Processor Scheduling n CPU scheduling more complex when multiple CPUs or multiple cores are

NUMA and CPU Scheduling Note that memory-placement algorithms can also consider affinity NUMA: non-uniform

Multiple-Processor Scheduling – Load Balancing n If SMP, need to keep all CPUs loaded

Real-Time CPU Scheduling n Can present obvious challenges n Soft real-time systems l No

Priority-based Scheduling n For real-time scheduling, scheduler must support preemptive, priority- based scheduling l

Rate Monotonic Scheduling n A priority is statically assigned based on the inverse of

Missed Deadlines with Rate Monotonic Scheduling n Example: P 1 is assigned a higher

Earliest Deadline First Scheduling (EDF) n Priorities are assigned according to deadlines l The

Proportional Share Scheduling n T shares are allocated among all processes in the system

End of Chapter 6 Operating System Concepts – 9 th Edition Silberschatz, Galvin and

Slides: 42

Download presentation

Chapter 6: CPU Scheduling n Basic Concepts n Scheduling Criteria n Scheduling Algorithms n Multiple-Processor Scheduling n Real-Time CPU Scheduling Operating System Concepts – 9 th Edition 6. 2 Silberschatz, Galvin and Gagne © 2013

Objectives n To introduce CPU scheduling, which is the basis for multiprogrammed operating systems n To describe various CPU-scheduling algorithms Operating System Concepts – 9 th Edition 6. 3 Silberschatz, Galvin and Gagne © 2013

Basic Concepts n Maximum CPU utilization obtained with multiprogramming n CPU–I/O Burst Cycle – Process execution consists of a cycle of CPU execution and I/O wait n CPU burst followed by I/O burst n CPU burst distribution is of main concern Operating System Concepts – 9 th Edition 6. 5 Silberschatz, Galvin and Gagne © 2013

Histogram of CPU-burst Times n Most often, we only look at one CPU-burst of a process when talking about scheduling algorithms Operating System Concepts – 9 th Edition 6. 6 Silberschatz, Galvin and Gagne © 2013

CPU Scheduler n Short-term scheduler selects from among the processes in ready queue, and allocates the CPU to one of them l Queue may be ordered in various ways n CPU scheduling decisions may take place when a process: 1. Switches from running to waiting state 4 2. Switches from running to ready state 4 l An interrupt occurs or the process finishes its time slot Switches from waiting to ready 4 l I/O request or wait() for child process Completion of I/O Terminates n Scheduling under 1 and 4 is nonpreemptive n All other scheduling is preemptive Operating System Concepts – 9 th Edition 6. 7 Silberschatz, Galvin and Gagne © 2013

Dispatcher n Dispatcher module gives control of the CPU to the process selected by the short-term scheduler; this involves: l switching context l switching to user mode l jumping to the proper location in the user program to restart that program n Dispatch latency – time it takes for the dispatcher to stop one process and start another running Operating System Concepts – 9 th Edition 6. 8 Silberschatz, Galvin and Gagne © 2013

Scheduling Criteria n CPU utilization – what percentage of CPU time is used for computation instead of idling n Throughput – # of processes that complete their execution per time unit n Turnaround time – amount of time to execute a particular process l The interval from the time of submission of a process to the time of completion n Waiting time – amount of time a process has been waiting in the ready queue n Response time – amount of time it takes from when a request was submitted until the first response is produced, not output (for time-sharing environment) Operating System Concepts – 9 th Edition 6. 10 Silberschatz, Galvin and Gagne © 2013

Scheduling Algorithm Optimization Criteria n Max CPU utilization n Max throughput n Min turnaround time n Min waiting time n Min response time Operating System Concepts – 9 th Edition 6. 11 Silberschatz, Galvin and Gagne © 2013

Chapter 6: CPU Scheduling n Basic Concepts n Scheduling Criteria n Scheduling Algorithms l First-come, first-served scheduling l Shortest-job-first scheduling l Priority scheduling l Round-robin scheduling l Multilevel queue scheduling l Multilevel feedback queue scheduling n Multiple-Processor Scheduling n Real-Time CPU Scheduling Operating System Concepts – 9 th Edition 6. 12 Silberschatz, Galvin and Gagne © 2013

First- Come, First-Served (FCFS) Scheduling Process Burst Time P 1 24 P 2 3 P 3 3 n Suppose that the processes arrive in the order at time 0: P 1 , P 2 , P 3 The Gantt Chart for the schedule is: n Waiting time for P 1 = 0; P 2 = 24; P 3 = 27 n Average waiting time: (0 + 24 + 27)/3 = 17 Operating System Concepts – 9 th Edition 6. 13 Silberschatz, Galvin and Gagne © 2013

FCFS Scheduling (Cont. ) Suppose that the processes arrive in the order at time 0: P 2 , P 3 , P 1 n The Gantt chart for the schedule is: n Waiting time for P 1 = 6; P 2 = 0; P 3 = 3 n Average waiting time: (6 + 0 + 3)/3 = 3 n Much better than previous case n Convoy effect - short process behind long process l Consider one CPU-bound and many I/O-bound processes Operating System Concepts – 9 th Edition 6. 14 Silberschatz, Galvin and Gagne © 2013

Shortest-Job-First (SJF) Scheduling n Associate with each process the length of its next CPU burst l Use these lengths to schedule the process with the shortest time n SJF is optimal – gives minimum average waiting time for a given set of processes l The difficulty is to know the length of the next CPU request l Could ask the user Operating System Concepts – 9 th Edition 6. 15 Silberschatz, Galvin and Gagne © 2013

Example of SJF Process. Arrival Time Burst Time P 1 0. 0 6 P 2 2. 0 8 P 3 4. 0 7 P 4 5. 0 3 n Four processes arrive at time 0 n SJF scheduling chart n Average waiting time = (3 + 16 + 9 + 0) / 4 = 7 Operating System Concepts – 9 th Edition 6. 16 Silberschatz, Galvin and Gagne © 2013

Determining Length of Next CPU Burst n Can only estimate the length – should be similar to the previous one l Then pick process with shortest predicted next CPU burst n Can be done by using the length of previous CPU bursts, using exponential averaging n Commonly, α set to ½ Operating System Concepts – 9 th Edition 6. 17 Silberschatz, Galvin and Gagne © 2013

Examples of Exponential Averaging n =0 n+1 = n l Recent history does not count n =1 l n+1 = tn l Only the actual last CPU burst counts n If we expand the formula, we get: n+1 = tn+(1 - ) tn -1 + … +(1 - )j tn -j + … l +(1 - )n +1 0 n Since both and (1 - ) are less than or equal to 1, each successive term has less weight than its predecessor Operating System Concepts – 9 th Edition 6. 19 Silberschatz, Galvin and Gagne © 2013

Example of Shortest-remaining-time-first n Now we add the concepts of varying arrival times and preemption to the analysis Process. Aarri Arrival Time. T Burst Time P 1 0 8 P 2 1 4 P 3 2 9 P 4 3 5 n Preemptive SJF Gantt Chart n Average waiting time = [(10 -1)+(17 -2)+(5 -3)]/4 = 26/4 = 6. 5 msec Operating System Concepts – 9 th Edition 6. 20 Silberschatz, Galvin and Gagne © 2013

Priority Scheduling n A priority number (integer) is associated with each process n The CPU is allocated to the process with the highest priority (smallest integer highest priority) l Preemptive l Nonpreemptive n SJF is priority scheduling where priority is the inverse of predicted next CPU burst time n Problem Starvation – low priority processes may never execute n Solution Aging – as time progresses increase the priority of the process Operating System Concepts – 9 th Edition 6. 21 Silberschatz, Galvin and Gagne © 2013

Example of Priority Scheduling Process. A arri Burst Time. T Priority P 1 10 3 P 2 1 1 P 3 2 4 P 4 1 5 P 5 5 2 n Five processes arrive at time 0 n Priority scheduling Gantt Chart n Average waiting time = 8. 2 msec Operating System Concepts – 9 th Edition 6. 22 Silberschatz, Galvin and Gagne © 2013

Round Robin (RR) n Each process gets a small unit of CPU time (time quantum q) l Usually 10 -100 milliseconds l After this time has elapsed, the process is preempted and added to the end of the ready queue. n If there are n processes in the ready queue and the time quantum is q l Each process gets 1/n of the CPU time in chunks of at most q time units at once l No process waits more than (n-1)q time units. n Timer interrupts every quantum to schedule next process n Performance l q large FIFO l q small q must be large with respect to context switch, otherwise overhead is too high Operating System Concepts – 9 th Edition 6. 23 Silberschatz, Galvin and Gagne © 2013

Example of RR with Time Quantum = 4 Process Burst Time P 1 P 2 P 3 n Three processes arrive at time 0 n The Gantt chart is: 24 3 3 n Typically, higher average turnaround than SJF, but better response n q should be large compared to context switch time n q usually 10 ms to 100 ms, context switch < 10 usec Operating System Concepts – 9 th Edition 6. 24 Silberschatz, Galvin and Gagne © 2013

Multilevel Queue n Ready queue is partitioned into separate queues, e. g. , l foreground (interactive) l background (batch) n Process permanently in a given queue n Each queue has its own scheduling algorithm, e. g. , l foreground – RR l background – FCFS n Scheduling must be done between the queues: l Fixed priority scheduling; (i. e. , serve all from foreground then from background). Possibility of starvation. l Time slice – each queue gets a certain amount of CPU time which it can schedule amongst its processes; 4 e. g. , 80% to foreground in RR, 20% to background in FCFS Operating System Concepts – 9 th Edition 6. 27 Silberschatz, Galvin and Gagne © 2013

Multilevel Feedback Queue n A process can move between the various queues; aging can be implemented this way n Multilevel-feedback-queue scheduler defined by the following parameters: l number of queues l scheduling algorithms for each queue l method used to determine when to upgrade a process l method used to determine when to demote a process l method used to determine which queue a process will enter when that process needs service Operating System Concepts – 9 th Edition 6. 29 Silberschatz, Galvin and Gagne © 2013

Example of Multilevel Feedback Queue n Three queues: l Q 0 – RR with time quantum 8 milliseconds l Q 1 – RR time quantum 16 milliseconds l Q 2 – FCFS n Scheduling l l A new job enters queue Q 0 which is served FCFS 4 When it gains CPU, job receives 8 milliseconds 4 If it does not finish in 8 milliseconds, job is moved to queue Q 1 At Q 1 job is again served FCFS and receives 16 additional milliseconds 4 If it still does not complete, it is preempted and moved to queue Q 2 Operating System Concepts – 9 th Edition 6. 30 Silberschatz, Galvin and Gagne © 2013

Multiple-Processor Scheduling n CPU scheduling more complex when multiple CPUs or multiple cores are available n Homogeneous processors within a multiprocessor n Asymmetric multiprocessing – only one processor accesses the system data structures, alleviating the need for data sharing n Symmetric multiprocessing (SMP) – each processor is self- scheduling, all processes in common ready queue, or each has its own private queue of ready processes l Currently, most common n Processor affinity – process has affinity for processor on which it is currently running l soft affinity 4 OS l attempts to keep a process running on the same processor hard affinity 4 Allow a process to specify a subset of processors on which it may run Operating System Concepts – 9 th Edition 6. 32 Silberschatz, Galvin and Gagne © 2013

NUMA and CPU Scheduling Note that memory-placement algorithms can also consider affinity NUMA: non-uniform memory access Operating System Concepts – 9 th Edition 6. 33 Silberschatz, Galvin and Gagne © 2013

Multiple-Processor Scheduling – Load Balancing n If SMP, need to keep all CPUs loaded for efficiency n Load balancing attempts to keep workload evenly distributed n Push migration – periodic task checks load on each processor, and if found pushes task from overloaded CPU to other CPUs n Pull migration – idle processors pulls waiting task from busy processor Operating System Concepts – 9 th Edition 6. 34 Silberschatz, Galvin and Gagne © 2013

Real-Time CPU Scheduling n Can present obvious challenges n Soft real-time systems l No guarantee as to when critical real-time process will be scheduled l Only guarantee that critical real-time process will be given preferences over noncritical processes n Hard real-time systems l Task must be serviced by its deadline 4 The moment before which the task has to be completed Operating System Concepts – 9 th Edition 6. 36 Silberschatz, Galvin and Gagne © 2013

Priority-based Scheduling n For real-time scheduling, scheduler must support preemptive, priority- based scheduling l But only guarantees soft real-time n Hard real-time systems must also provide ability to meet deadlines n Processes have new characteristics: periodic ones require CPU at constant intervals l Has processing time t, deadline d, period p l 0≤t≤d≤p l Rate of periodic task is 1/p Operating System Concepts – 9 th Edition 6. 37 Silberschatz, Galvin and Gagne © 2013

Rate Monotonic Scheduling n A priority is statically assigned based on the inverse of its period l Shorter periods = higher priority l Longer periods = lower priority n Example: P 1 is assigned a higher priority than P 2 l p 1=50, p 2=100 l t 1=20, t 2=35 l Deadline of both: the start of the next cycle Operating System Concepts – 9 th Edition 6. 38 Silberschatz, Galvin and Gagne © 2013

Missed Deadlines with Rate Monotonic Scheduling n Example: P 1 is assigned a higher priority than P 2 l p 1=50, p 2=80 l t 1=25, t 2=35 l Deadline of both: the start of the next cycle Operating System Concepts – 9 th Edition 6. 39 Silberschatz, Galvin and Gagne © 2013

Earliest Deadline First Scheduling (EDF) n Priorities are assigned according to deadlines l The earlier the deadline, the higher the priority l The later the deadline, the lower the priority l Processes don’t have to be periodic any more 4 A process has to announce its deadline n Example: P 1 and P 2 l p 1=50, p 2=80 l t 1=25, t 2=35 l Deadline of both: the start of the next cycle Operating System Concepts – 9 th Edition 6. 40 Silberschatz, Galvin and Gagne © 2013

Proportional Share Scheduling n T shares are allocated among all processes in the system n An application receives N shares where N < T n This ensures each application will receive N / T of the total processor time n Example: T = 100 shares l Process A: 50 shares l Process B: 15 shares l Process C: 20 shares Operating System Concepts – 9 th Edition 6. 41 Silberschatz, Galvin and Gagne © 2013