We are here to give you the Sight & Sound of Information Technology

Scheduling Criteria


493
Scheduling Criteria

Different scheduling algorithms have different properties and functionality. The choice of a particular algorithm may favor a particular class of processes. While choosing an algorithm for CPU scheduling, we must consider some properties of algorithms. We can compare the CPU scheduling algorithms on the basis of following criteria:

CPU utilization:

In general, we want our CPU to be busy all the time. Conceptually, the CPU utilization may range from 0 to 100 percent. But in real systems, the CPU utilization may range from 40 to 90 percent. 40 percent indicates a CPU which is loaded lightly while 90 percent indicates a CPU which is used heavily. The CPU utilization must always be optimal.

Throughput:

The total number of processes that are completed in a unit time, is called throughput. If CPU is continuously busy in executing the processes, then definitely work will be done by CPU. If long processes are being executing, then throughput will be low. But if short processes are being executing, the throughput will be high. We always prefer an algorithm whose throughput is high.

Turn-around Time:

The turnaround time shows the total life span of a process. The turnaround time will be equal to the time taken by a process from its arrival till its termination. The turnaround time is also described as the interval between time of submission of a process and its time of completion. We always prefer an algorithm which reduces the turnaround time of each process. Following is the formula for calculating turnaround time:

Turnaround time = Time of execution completion - arrival time.

Waiting Time:

It is the total time which is spent by a process in waiting in ready queue. The different CPU scheduling algorithms causes different waiting time for different processes. We always prefer an algorithm which reduces the waiting time of each process. Following is the formula for calculating turnaround time:

Waiting Time = Completion time - total time required by a process on CPU - Arrival time of a process.

Response Time:

The time interval between the submission of a process and its first response (execution), is called response time.

Optimal Conditions For selecting an algorithm:

We always select an algorithm which will:

  • Increase the CPU utilization.
  • Increase the Throughput.
  • Reduce the Turnaround time.
  • Reduce the Waiting Time.
  • Reduce the Response time.

In most cases, we use the algorithm whose average of all five criteria is good. For time sharing systems, we prefer to minimize the variance in response time rather than minimizing the average response time. A system which has stable response time is far better than a system which has less average response time but is highly variable.

Tags

Copy Right

The contents available on this website are copyrighted by TechPlus unless otherwise indicated. All rights are reserved by TechPlus, and content may not be reproduced, published, or transferred in any form or by any means, except with the prior written permission of TechPlus.