A model for an adaptive scheduler

Authors:
Carlos Gonzalez;Kian Tavakoli
Affiliations:
Computer Science Department, School of Information Technology and Engineering, George Mason University, Fairfax, Va;Computer Science Department, School of Information Technology and Engineering, George Mason University, Fairfax, Va
Venue:
CSC '88 Proceedings of the 1988 ACM sixteenth annual conference on Computer science
Year:
1988

Citing 5
Cited 0

A distributed load-balancing policy for a multicomputer

Software—Practice & Experience
Effects of job loading policies for multiprogramming systems in processing a job stream

ACM Transactions on Computer Systems (TOCS)
Scheduling with deadline requirements

ACM '85 Proceedings of the 1985 ACM annual conference on The range of computing : mid-80's perspective: mid-80's perspective
Maximum Processing Rates of Memory Bound Systems

Journal of the ACM (JACM)
Optimality of scheduling policy for processing a job stream

ACM Transactions on Computer Systems (TOCS)

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper reports on the design of a model for an adaptive scheduler of an operating system. The model is based on an adaptive control loop system in which the system's performance quality is evaluated by an index of performance, defined here as the users dissatisfaction time. At each user interaction, the system performs three basic functions: a) select the expected service type of user interaction based on a history of past interactions, b) assign a deadline for the system to respond to the user interaction, and c) modify (adapt) some parameters used in calculating the user's priority (using the system's history of previous services to the user) to reduce the probability of the user being dissatisfied in this interaction. We present here the algorithms necessary to implement this model. Numerical results from a simulation show that our adaptive scheduler is more efficient in minimizing the number of dissatisfied users than any other previously known method.