Chapter 5: Process Scheduling

Chapter 5: Process SchedulingChapter 5: Process Scheduling

5.2 Silberschatz, Galvin and Gagne ©2005Operating System Concepts

Basic ConceptsBasic Concepts

CPU scheduling is the basic of multiprogramming OS.

Scheduling is a fundamental OS function.

Almost all computer resources are scheduled before use.

The CPU is one of the primary computer resources.

By switching the CPU among processes, the OS can make the computer more productive.

Thus its scheduling is central to OS design


CPU-I/O Burst Cycle CPU-I/O Burst Cycle

CPU–I/O Burst Cycle – Process execution consists of a cycle of CPU execution and I/O wait

Processes alternate between these two states.

Process execution begins with a CPU burst.

That is followed by an I/O burst, then another CPU burst, then another I/O burst and so on.


CPU-I/O Burst CycleCPU-I/O Burst Cycle

Finally, the last CPU burst will end with a system request to terminate execution , rather than with another I/O burst.

This distribution can help us select an appropriate CPU-scheduling algorithm.


Alternating Sequence of CPU And I/O BurstsAlternating Sequence of CPU And I/O Bursts


CPU-I/O Burst CycleCPU-I/O Burst Cycle

The duration of these CPU burst have been extensively measured.

In measured in histogram the curve is generally characterized as exponential or hyper exponential with many short CPU bursts, and a few long CPU bursts.


Histogram of CPU-burst TimesHistogram of CPU-burst Times


CPU SchedulerCPU Scheduler

Selects from among the processes in memory that are ready to execute, and allocates the CPU to one of them

CPU scheduling decisions may take place when a process:

1.Switches from running to waiting state

2. Switches from running to terminates state

3. Switches from running to ready state

3.Switches from waiting to ready state


Preemptive versus Non-preemptive Preemptive versus Non-preemptive SchedulingScheduling

Non-Preemptive scheduling : Scheduling takes place only under first

two of the above cases. Here, one process is selected to execute.

Then, it is allowed to run until it voluntarily enter into the wait state (case 1) or gets terminate (case 2).

Simple, but its not efficient. doesn’t require timer or clock interrupt.


Preemptive versus Non-preemptive Preemptive versus Non-preemptive SchedulingScheduling

Preemptive scheduling :

Scheduling takes place only under any of the above cases.

Here, one process is selected to execute. Then, it is allowed to run only for some maximum time duration. After the time duration, another process is selected to execute.

Very efficient, but complex.

require timer or clock interrupt.


DispatcherDispatcher

Dispatcher module gives control of the CPU to the process selected by the short-term scheduler; this function involves:

switching context

switching to user mode

jumping to the proper location in the user program to restart that program

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


Scheduling CriteriaScheduling Criteria

Various scheduling algorithms are available.

They work based on different criteria to select a process.

To compare the performance of different scheduling algorithms, various performance criteria are decided.



They are as given below : CPU utilization :

It is an average fraction of time during which CPU is busy.

It ranges form 0 to 100 percent. In a real system, it should be between 40 to 90 percent.

CPU should remain as busy as possible, or CPU utilization should remain as high as possible.



Throughput : Number of processes completed per time unit is

called throughput. For long process, this rate may be 1 hour; for

short transactions, throughput might be 10 processes per second.

Turnaround time : Time required to complete execution of a process

is called turn-around time. It specifies that how long it takes to complete a

process execution. Turnaround time=process finish time-process

arrival time.



Waiting time : It is total time duration spent by a

process waiting in ‘Ready’ queue. Waiting time=turnaround time-actual

execution time. Scheduling algorithm affects the time

that a process spends waiting in the ‘Ready’ queue.



Response time : It is time between issuing a

command/request and getting output/result.

It is more important in interactive system.

User requests should be served as quickly as possible.

Here, user requests are given more priority compared to background system processes.


Optimization CriteriaOptimization Criteria

Max CPU utilization

Max throughput

Min turnaround time

Min waiting time

Min response time


Scheduling AlgorithmsScheduling Algorithms

There are four different algorithms are described bellow :

First-Come, First-Served (FCFS) Scheduling

Shortest-Job-First (SJF) Scheduling

Priority Scheduling

Round-Robin Scheduling


First-Come, First-Served (FCFS) SchedulingFirst-Come, First-Served (FCFS) Scheduling

Selection criteria : The process that requests first is served

first. It means that processes are served in the exact order as they come.

Decision Mode: Non-Preemptive: once a process is

selected, it runs until it blocks for an I/O or some event, or it terminates.



Implementation:

This strategy can be easily implemented by using FIFO queue.

FIFO means first in first out. When first process enters the system, it starts its execution.

All other processes are appended in a queue. When CPU becomes free, a process from the first position in a queue is selected to run.



Example:

Consider the following set of four processes. Their arrival time and time required to complete the execution are given in following table.

Consider all time values in milliseconds.



Process Arrival Time Time required for

Completion

(Burst time)

P0 0 10

P1 1 6

P2 3 2

P3 5 4



Gantt Chart:

Average Turnaround Time:(10+15+15+17)/4=57/4=14.25 ms

Average Waiting Time:(0+9+13+13)/4=35/4=8.75 ms

0 10 16 18 2218 221816 221810 16 22180 10 16 2218

P0 P1 P2 P3



Initially only process P0 is present and it is allowed to run.

But, when P0 completes, all other processes are present.

So, next process P1 from ready queue is selected and allowed to run till it completes.

This procedure is repeated till all processes complete their execution.



Advantages:

Simple, Fair

Easy to understand, easy to implement

Disadvantages:

Not efficient. Average waiting time is too high.

CPU utilization may be too low. Consider a CPU-bound process running with many I/O bound process.


Shortest-Job-First (SJR) SchedulingShortest-Job-First (SJR) Scheduling

Associate with each process the length of its next CPU burst. Use these lengths to schedule the process with the shortest time

Two schemes:

Non-Preemptive – once CPU given to the process it cannot be preempted until completes its CPU burst


Shortest-Job-First (SJF) SchedulingShortest-Job-First (SJF) Scheduling

Preemptive – if a new process arrives with CPU burst length less than remaining time of current executing process, preempt. This scheme is know as the Shortest-Remaining-Time-First (SRTF)

SJF is optimal – gives minimum average waiting time for a given set of processes


Shortest-Job-First (SJF) SchedulingShortest-Job-First (SJF) SchedulingNon-preemptiveNon-preemptive

Selection Criteria:

The process that requires shortest time to complete execution, is served first.

Decision Mode:

Non-preemptive: Once a process is selected, it runs until it blocks for an I/O or some event, or it terminates.



Implementation:

This strategy can be implemented by using sorted FIFO queue.

All processes in a queue are sorted in ascending order based on their required CPU burst.

When CPU becomes free, a process from the first position in queue is selected to run.



Example:






Completion

(Burst time)

P0 0 10

P1 1 6

P2 3 2

P3 5 4



Gantt Chart:

Average Turnaround Time:

(10+21+9+11)/4=51/4=12.75 ms Average Waiting Time:

(0+15+7+7)/4=29/4=7.25 ms

0 10 12 2216

P0 P2 P3 P1





So, process with shortest CPU burst P2 is selected and allowed to run till it completes.

Whenever more than one process is available, such type of decision is taken.

This procedure is repeated till all process complete their execution.



Advantages:

Less waiting time.

Good response for short processes

Disadvantages:

It is difficult to estimate time required to complete execution.

Starvation is possible for long process. Long process may wait for ever.


Shortest-Job-First (SJF) SchedulingShortest-Job-First (SJF) SchedulingPreemptivePreemptive

Selection Criteria:

The process, whose remaining run time is shortest, is served first.

This is a preemptive version of SJF scheduling

Decision Mode:

Preemptive: when a new process arrives, it total time is compared to the current process remaining time.


Shortest-Job-First (SJF) Scheduling Shortest-Job-First (SJF) Scheduling PreemptivePreemptive

If the new job needs less time to finish than the current process, the current process is suspended and new job started.

Implementation: This strategy can be implemented by using

sorted FIFO queue. All processes in a queue are sorted in

ascending order based on their remaining run time.




Example:






Completion

(Burst time)

P0 0 10

P1 1 6

P2 3 2

P3 5 4



Gantt Chart:



(12+2+0+4)/4=18/4=4.5 ms

0 3 5 2213

P0 P2 P1 P3P0 P1 P0

91

P1 P2P1 P2P1 P2P1 P2P1 P0P2 P0P2




But, when P1 comes, it has shortest remaining run time.

So, P0 is preempted and P1 is allowed to run.

Whenever new process comes or current process blocks, such type of decision is taken.

This procedure is repeated till all process complete their execution.



Advantages:

Less waiting time.

Good response for short processes

Disadvantages:

Starvation is possible for long process. Long process may wait for ever.

Context switch overhead is there.


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

The CPU is allocated to the process with the highest priority (smallest integer highest priority)

Preemptive

Non-Preemptive

SJF is a priority scheduling where priority is the predicted next CPU burst time

Problem Starvation – low priority processes may never execute

Solution Aging – as time progresses increase the priority of the process


Priority Based SchedulingPriority Based SchedulingNon-PreemptiveNon-Preemptive

Selection Criteria:

The process, the has highest priority is served first.

Decision Mode:

Non-preemptive: Once a process is selected, it runs until it blocks for an I/O or some event, or it terminates.


Priority Based SchedulingPriority Based Scheduling Non-Non-PreemptivePreemptive


sorted FIFO queue. All processes in a queue are sorted based

on their priority with highest priority process at front end.



Priority Based Scheduling Priority Based Scheduling Non-PreemptiveNon-Preemptive

Example:

Consider the following set of four processes. Their arrival time and time required to complete the execution and priorities are given in following table.

Consider all time values in milliseconds and small value for priority means higher priority of a process.



Process Arrival Time

Time required for

Completion

(Burst time)

Priority

P0 0 10 5

P1 1 6 4

P2 3 2 2

P3 5 4 0



Gantt Chart:



(0+15+11+5)/4=31/4=7.75 ms

0 2216

P3 P2 P1

1410

P0





So, process with highest priority P3 selected and allowed to run till it completes.

Whenever more than one process is available, such type of decision is taken.




Advantages:

Priority is considered. Critical processes can get better response time.

Disadvantages:

Starvation is possible for low priority processes. It can be overcome by using technique called ‘Aging’.

Aging : gradually increases the priority of processes that wait in the system for a long time.


Priority Based SchedulingPriority Based SchedulingPreemptivePreemptive

Selection Criteria:

The process, the has highest priority is served first.

Decision Mode:

Preemptive: When a new process arrives, its priority is compared to the current process priority.

If the new job has higher priority than the current process, the current process is suspended and the new job started.


Priority Based SchedulingPriority Based Scheduling PreemptivePreemptive


sorted FIFO queue. All processes in a queue are sorted based

on their priority with highest priority process at front end.



Priority Based Scheduling Priority Based Scheduling PreemptivePreemptive

Example:

Consider the following set of four processes. Their arrival time and time required to complete the execution and priorities are given in following table.

Consider all time values in milliseconds and small value for priority means higher priority of a process.



Process Arrival Time

Time required for

Completion

(Burst time)

Priority

P0 0 10 5

P1 1 6 4

P2 3 2 2

P3 5 4 0



Gantt Chart:



(12+6+0+0)/4=18/4=4.5 ms

0 221391

P0P2P1P0 P1 P1

3 5




But, P1 comes, it has higher priority.

So, P0 is preempted and P1 is allowed to run.




Advantages:

Priority is considered. Critical processes can get better response time.

Disadvantages:

Starvation is possible for low priority processes. It can be overcome by using technique called ‘Aging’.

Aging : gradually increases the priority of processes that wait in the system for a long time.



Round Robin (RR)Round Robin (RR)

Each process gets a small unit of CPU time (time quantum), usually 10-100 milliseconds. After this time has elapsed, the process is preempted and added to the end of the ready queue.

If there are n processes in the ready queue and the time quantum is q, then each process gets 1/n of the CPU time in chunks of at most q time units at once. No process waits more than (n-1)q time units.



Selection Criteria:

The Process that requests first is serves first.

It means that Process are served in the exact order as they come.

The selection criteria are same as that of FCFS scheduling.



Decision Mode: Non-Preemptive/Preemptive : Each

selected process is assigned a time interval, called time quantum or time slice.

Process is allowed to run only for this time interval.

Here, one of two things is possible : First, process needs CPU burst less than time quantum.



In this case, process will voluntarily release CPU burst less than to the end of the queue.

Second, process needs CPU burst longer than time quantum.

In this case, process will be running at the end of time quantum.

Now, it will be preempted and moved to the end of the queue.

CPU will be allocated to another process. Here, length of time quantum is critical to

determine.




sorted FIFO queue. When new Process, or process releases

CPU, or process is preempted, it is moved to the end of the queue.




Example: (Non-Preemptive)

Consider the following set of four processes.

Their burst time are given in following table.

all time values are in milliseconds.

Consider that time quantum is of 4 millisecond.



Process Burst

Time

P0 24

P1 3

P2 3



Gantt Chart:


(30+7+10)/3=47/3=15.66 ms Average Waiting Time:

(6+4+7)/3=17/3=5.56 ms

0 30264

P1 P1

14 18

P1 P3P2P2 P3 P1 P1 P1 P1 P1P1

7 10 22



When we use a time quantum of 4 ms, then process P1 gets the first 4 ms.

Since, it requires another 20 ms, it is preempted after the first time quantum, and the CPU is given to the next process in the queue, process P2.

Since, process P2 does not need 4 ms, it quits before its time quantum expires.

The CPU is the given to the next process, process P3.

Once each process has received 1 time quantum, the CPU is returned to process P1 for an additional time quantum.



Advantages:

One of the oldest, simplest, fairest and most widely used algorithms.

Disadvantages:


Determination of time quantum is too critical. If it is too short, it causes frequent context switches and lowers CPU efficiency. If it is too long it causes poor responses for short interactive processes.


Time Quantum and Context Switch TimeTime Quantum and Context Switch Time


Multilevel Queue SchedulingMultilevel Queue Scheduling

Ready queue is partitioned into separate queues:

foreground (interactive)background (batch)

Each queue has its own scheduling algorithm

foreground – RR

background – FCFS



Scheduling must be done between the queues Fixed priority scheduling; (i.e., serve all

from foreground then from background). Possibility of starvation.

Time slice – each queue gets a certain amount of CPU time which it can schedule amongst its processes; i.e., 80% to foreground in RR

20% to background in FCFS


Multilevel Feedback Queue SchedulingMultilevel Feedback Queue Scheduling

A process can move between the various queues; aging can be implemented this way

Multilevel-feedback-queue scheduler defined by the following parameters:

number of queues

scheduling algorithms for each queue

method used to determine when to upgrade a Process


Multilevel Feedback Queue SchedulingMultilevel Feedback Queue Scheduling

method used to determine when to demote a process

method used to determine which queue a process will enter when that process needs service


Example of Multilevel Feedback QueueExample of Multilevel Feedback Queue

Three queues:

Q0 – RR with time quantum 8 milliseconds

Q1 – RR time quantum 16 milliseconds

Q2 – FCFS


Example of Multilevel Feedback QueueExample of Multilevel Feedback Queue

Scheduling A new job enters queue Q0 which is

served FCFS. When it gains CPU, job receives 8 milliseconds. If it does not finish in 8 milliseconds, job is moved to queue Q1.

At Q1 job is again served FCFS and receives 16 additional milliseconds. If it still does not complete, it is preempted and moved to queue Q2.


Multilevel Feedback QueuesMultilevel Feedback Queues


Multiple-Processor SchedulingMultiple-Processor Scheduling

CPU scheduling more complex when multiple CPUs are available

Homogeneous (Identical) processors within a Multiprocessor.

If several identical processors are available, then Load sharing can occur.

It would be possible to provide a separate queue for each processor.

Asymmetric multiprocessing – only one processor accesses the system data structures, alleviating the need for data sharing.


Real-Time SchedulingReal-Time Scheduling

Hard real-time systems – required to complete a critical task within a guaranteed amount of time

Soft real-time computing – requires that critical processes receive priority over less fortunate ones


Thread SchedulingThread Scheduling

Local Scheduling – How the threads library decides which thread to put onto an available lightweight process(LWP).

Global Scheduling – How the kernel decides which kernel thread to run next.


Java Thread SchedulingJava Thread Scheduling

JVM Uses a Preemptive, Priority-Based Scheduling Algorithm

FIFO Queue is Used if There Are Multiple Threads With the Same Priority


Java Thread Scheduling (cont)Java Thread Scheduling (cont)

JVM Schedules a Thread to Run When

1. The Currently Running Thread Exits the Runnable State

2. A Higher Priority Thread Enters the Runnable State


Operating System ExamplesOperating System Examples

Solaris scheduling

Windows XP scheduling

Linux scheduling


Solaris 2 SchedulingSolaris 2 Scheduling

Solaris 2 uses priority-based process scheduling.

It has four classes of scheduling, which are in order of priority real time, system, time sharing and interactive.

Each class includes different priorities and scheduling algorithms.



Time sharing class:

The scheduling policy for this class to dynamically alters priorities and assign time slices of different lengths using a multilevel feedback queue.

Interactive class :

This class uses the same scheduling policy as the time-sharing class, but it gives windowing application a higher priority for better performance.



System class:

Solaris 2 uses the system class to run kernel processes,.

Once established, the priority of a system process does not change.

The system class reserved for kernel use.

The scheduling policy for the system class does not time-slice.



Real time class:

This class is given the highest priority to run among all classes.

So a real time process will run before a process in any other class.


Windows XP SchedulingWindows XP Scheduling

Windows XP schedules threads using a priority-based, preemptive scheduling algorithm.

The thread selected to run by the dispatcher will run until it is preempted by a higher-priority thread until it terminates, until its time quantum ends or until it calls a blocking system call, such as for I/O.

The dispatcher uses a 32-level priority scheme to determine the order of thread execution.



Priorities are divided into two classes: the variable class contains threads having priorities from 1 to 15.

And the real time class contains thread with priorities ranging from 16 to 31.

There is also thread running at priority 0 that is used for memory management.



There is a relationship between the numeric priority of the windows XP kernel and the win32 API.

In next slide figure to shown the values of each priority class appear in the top row.

The left column contains the values for the different relative priorities.

For example , if the relative priority of a thread in the ABOVE_NORMAL_PRIORITY_CLASS is NORMAL, the numeric priority of that thread is 10.


Windows XP PrioritiesWindows XP Priorities


Linux SchedulingLinux Scheduling

Linux provides two separate process-scheduling algorithms. Time-sharing real-time



Time-sharing :

Linux uses a prioritized, credit-based algorithm.

Each process possesses a certain number of scheduling credits, when a new task must be chosen to run, the process with the most credits is selected.

Every time that a time interrupt occurs, the currently running process loses one credit, when its credits reaches zero , it is suspended and another process is chosen.



real-time :

Linux’s real-time scheduling is soft-rather than hard- real time.

The scheduler offers strict guarantees about the relative priorities of real-time processes, but the kernel does not offer any guarantees about how quickly a real time process will be scheduled once that process becomes runble.



Real-time Soft real-time Posix.1b compliant – two classes

FCFS and RRHighest priority process always runs

first


Algorithm EvaluationAlgorithm Evaluation

Deterministic modeling

takes a particular predetermined workload and defines the performance of each algorithm for that workload

Queuing models

Queuing analysis can useful in comparing scheduling algorithms but it also has limitations.

The mathematics of complicated algorithms or distribution can be difficult to work with.


Algorithm EvaluationAlgorithm Evaluation

So that they will be able to compute an answer, queuing models are often only an approximation of real system.

Implementation

The only completely accurate way to evaluate a scheduling algorithm is to code it, put it in the operating system and see how it works.

This approach puts the actual algorithm in the real system for evaluation under real operating condition.

End of Chapter 5End of Chapter 5

Chapter 5: Process Scheduling

Documents

Transcript of Chapter 5: Process Scheduling