CPU Scheduling. CPU Scheduling

ECE/IUPUI RTOS & APPS 1

CPU Scheduling

Electrical and Computer Engineering Stephen Kim ([email protected])

CPU Scheduling

n

Basic Concepts

n

Scheduling Criteria

n

Scheduling Algorithms

n

Multiple-Processor Scheduling

n

Real-Time Scheduling

n

Algorithm Evaluation

ECE/IUPUI RTOS & APPS 3

Basic Concepts

n Maximum CPU utilization obtained with

multiprogramming by making some process running at all times.

n CPU Burst vs. I/O Burst – Process execution consists of a cycle of CPU execution and I/O wait.

n CPU burst distribution

Alternating Sequence of CPU And I/O Bursts

CPU Burst IO Burst

IO Request

IO Complete

ECE/IUPUI RTOS & APPS 5

Histogram of CPU-burst Times

n

exponential or hyperexponential

n

IO bound program – many very short CPU burst eg.)

n

CPU bound program – a few very long CPU burst

CPU Scheduler

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

n CPU scheduling decisions may take place when a process:

1. Switches from running to waiting state.

2. Switches from running to ready state.

3. Switches from waiting to ready.

4. Terminates.

n Scheduling under 1 and 4 is nonpreemptive. A new ready process must be selected for execution.

n All other scheduling is preemptive.

ECE/IUPUI RTOS & APPS 7

Nonpreemptive vs. Preemptive

n Nonpreemptive

u A process keeps the CPU until it release the CPU either by terminating, or by switching to the waiting state.

u eg) MS Windows 3.1, Apple Macintosh

t limited by hardware support.

n Preemptive

u A process can be interrupted any time.

u Possible Inconsistency

t A process can be interrupted while it modifies some data, which other process tries to modify or read.

t Frequent between a system call and a user process.

t A simple solution is to disallow to interrupt in a kernel mode.

t The solution is not adequate for real-time systems.

Dispatcher

n

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

u switching context

u switching to user mode

u 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.

ECE/IUPUI RTOS & APPS 9

Scheduling Criteria

n CPU utilization – keep the CPU as busy as possible. A reasonable range is from 40% to 90%.

n Throughput – # of processes that complete their execution per time unit.

n Turnaround time – amount of time to execute a particular process from submission to 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).

Optimization Criteria

n

What we want to

u Maximize CPU utilization

u Maximize throughput

u Minimize turnaround time

u Minimize waiting time

u Minimize response time

n

In most cases

u optimize the average measure

n

In the following, we will use the average waiting

time

ECE/IUPUI RTOS & APPS 11

First-come, First-served (FCFS) Scheduling

Process Burst Time

P₁ 24

P₂ 3

P₃ 3

n Suppose that the processes arrive in the order: P₁, P₂, P₃ The Gantt Chart for the schedule is:

n Waiting time for P₁= 0; P₂= 24; P₃ = 27

n Average waiting time: (0 + 24 + 27)/3 = 17

P₁ P₂ P₃

24 27 30

0

FCFS Scheduling (Cont.)

Suppose that the processes arrive in the order P

, p

, p

.

n

The Gantt chart for the schedule is:

n

Waiting time for P

= 6; P

= 0

P

= 3

n

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

n

Much better than previous case.

n

Convoy effect: short processes behind long process

P₁ P₃

P₂

6

3 30

0

ECE/IUPUI RTOS & APPS 13

Shortest-Job-First (SJR) Scheduling

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

n Two schemes:

u nonpreemptive – once CPU given to the process, it cannot be preempted until completes its CPU burst.

u preemptive – if a new process arrives with CPU burst length less than remaining time of current executing process, preempt.

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

Process Arrival Time Burst Time

P₁ 0.0 7

P₂ 2.0 4

P₃ 4.0 1

P₄ 5.0 4

n SJF (non-preemptive)

n Average waiting time = (0 + 6 + 3 + 7)/4 = 4

Example of Non-Preemptive SJF

P₁ P₃ P₂

7

3 16

0

P₄

8 12

ECE/IUPUI RTOS & APPS 15

Example of Preemptive SJF

Process Arrival Time Burst Time

P₁ 0.0 7

P₂ 2.0 4

P₃ 4.0 1

P₄ 5.0 4

n SJF (preemptive)

n Average waiting time = (9 + 1 + 0 +2)/4 = 3

n Even the SJF is optimal, it cannot be implemented because we cannot see future.

P₁ P₂ P₃ P₄

t

Predicting Length of Next CPU Burst

n

Can only estimate the length.

n

Can be done by using the length of previous CPU bursts, using exponential averaging.

1. t_n= actual length of n^th CPU burst

2. τ_n+1 = predicted value for the next CPU burst 3. 0≤α≤1

4. Define:

τ_n+1= αt_n+ (1-α) τ_n

ECE/IUPUI RTOS & APPS 17

Predicting Length of Next CPU Burst, cont

Examples of Exponential Averaging

n α =0

u τn+1= τn

u Recent history does not count.

n α =1

u τn+1= t_n

u Only the actual last CPU burst counts.

n If we expand the formula, we get:

τn+1= αt_n+(1 -α) αt_n-1 + … +(1 -α)^j αt_n-1 + … +(1 -α)ⁿ⁼¹ t_nτ0

n Since both α and (1 - α) are less than or equal to 1, each successive term has less weight than its predecessor.

ECE/IUPUI RTOS & APPS 19

Implementation Consideration

n The coefficient of exponential average is a floating number

n The computation of floating number is burden in designing a kernel.

n Even worse, many microprocessor used for real-time operating systems do not have floating point unit (FPU).

n What we can do for that type of processors?

Priority Scheduling

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

u UNIX: -20 (highest) to 19 (lowest), default is 10

u microC/OS-II: 0 (highest) to 63 (lowest)

n The CPU is allocated to the process with the highest priority

u Preemptive

u nonpreemptive

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

n Starvation – low priority processes may never execute.

u Solution: Aging – as time progresses increase the priority of the process.

ECE/IUPUI RTOS & APPS 21

Round Robin (RR)

n Each process gets a small unit of CPU time (time quantum or time slice), usually 10-100 milliseconds. 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, 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.

n Performance

u q large ⇒FIFO

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

Example of RR with Time Quantum = 4

Process Burst Time

P₁ 24

P₂ 3

P₃ 3

The Gantt chart is:

n Average waiting time of RR = (6+4+7)/3=5.66

n Average waiting time of SJF = (6+0+3)/3=3

n The average waiting time of RR is often quite long.

P₁ P₂ P₃ P₁ P₁ P₁ P₁ P₁ P₁

P₂ P₃ P₁ P₁ P₁ P₁ P₁ RR

SJF

ECE/IUPUI RTOS & APPS 23

Time Quantum and Context Switch Time

n The average turnaround of RR is longer than SJF because CPU bursts of all processes interleave.

n But, better response.

Turnaround Time Varies With The Time

Quantum

ECE/IUPUI RTOS & APPS 25

Multilevel Queue

n Ready queue is partitioned into separate queues:

u foreground (interactive)

u background (batch)

n Each queue has its own scheduling algorithm,

u foreground – RR

u background – FCFS

n Scheduling must be done between the queues.

u Fixed priority scheduling; (i.e., serve all from foreground then from background). Possibility of starvation.

u Time slice – each queue gets a certain amount of CPU time which it can schedule amongst its processes; i.e.,

t 80% to foreground in RR

t 20% to background in FCFS

Is this true?

Multilevel Queue Scheduling

ECE/IUPUI RTOS & APPS 27

Multilevel Feedback Queue Scheduling

n A process can move between the various queues

u If a process uses to much CPU time, it will be moved to a lower- priority queue, but leaves IO-bound and interactive processes in the higher priority queue.

u Aging can be implemented with the same way.

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

u number of queues

u scheduling algorithms for each queue

u method used to determine when to upgrade a process

u method used to determine when to demote a process

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

Example of Multilevel Feedback Queue

Three queues:

u Q₀– time quantum 8 milliseconds

u Q₁– time quantum 16 milliseconds

u Q₂– FCFS n Scheduling

u A new job enters queue Q₀which 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.

u At Q₁job is again served FCFS and receives 16 additional milliseconds. If it still does not complete, it is preempted and moved to queue Q₂.

ECE/IUPUI RTOS & APPS 29

Multilevel Feedback Queues

Example, Scheduling

n For the given processes, answer the following question for FCFS, SJF, RR(q=1), RR(q=4), 3Level Feedback(q=1), and 3Level Feedback(q=1&2).

n What is the average waiting times for each scheduling?

n What is the turnaround time for each scheduling?

5 12

E

5 9

D

2 3

C

5 1

B

3 0

A

Processing Time Arrival Time

Prcess Name

ECE/IUPUI RTOS & APPS 31

Real-time Scheduling

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

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

n Priority inversion

u When a higher-priority process needs to access kernel data currently being accessed by another lower-priority process. The higher-priority process must wait for the lower-priority process to finish

u Solution: priority inheritance scheme

all processes accessing resources that a higher-priority process needs, inherit the high priority until they are done with the resources. When they are completed the resource access, their priority reverts to its original value

Dispatch Latency

n The conflict phase consists of

Preemption of any process running in the kernel

Release by low-priority processes resources needed by the high-priority process.

n If the preemption is not allowed, the dispatch latency is quite long, so most of real-time OS’s support preemption.

ECE/IUPUI RTOS & APPS 33

Rate Monotonic Scheduling (RMS)

n

In an RT system, how can we assign priorities to critical tasks?

n

Rate monotonic scheduling

u Task priorities are assigned based on how often tasks executes

u Tasks with the highest rate of execution are given the highest priority

u Assumption

t All tasks are periodic

t Tasks do not synchronize with one another, share resouces, or exchange data

t OS supports preemption

RMS Theorem

n

Given n tasks, all task hard real-time deadlines are always met if the inequality hold.

) 1 2 (

−

∑ ≤

i

i

n

T E

where E_i is the maximum execution time of task i, and T_i is the execution period of task i. Note that E_i is always smaller than T_iand E_i/T_icorresponds to the fraction of CPU time required to execute task i.

eg) 2(2^1/2-1)=0.828, 100(2^1/100-1)=0.695

If the system has 100 tasks, the total CPU utilization must be less than 69.5% to meet the real-time deadline.

ECE/IUPUI RTOS & APPS 35

n (2 ^1/n -1)

0 0.2 0.4 0.6 0.8 1 1.2

0 5 10 15 20 25 30 35

Example, RMS

n

Consider three periodic tasks, Task P

: E

=40, T

=150, Task P

: E

=100, T

=350, and Task P

: E

=20, T

=100

n

What priorities will you assign to each task from 1 to 3? (The smaller number, the higher priority)

n

Can the tasks be scheduled in real-time?