Performance evaluation through job scheduler modeling

  • Authors:
  • M. K. Rajaraman

  • Affiliations:
  • The University of Texas at Arlington, Arlington, Texas

  • Venue:
  • SCORE '82 Selected papers of the 1982 ACM SIGMETRICS workshop on Software Metrics: part 1
  • Year:
  • 1982

Quantified Score

Hi-index 0.01

Visualization

Abstract

The scheduler in the Cyber-176 computer does the major functions of routing the job through the system, controlling job's progress through aging and swapping of jobs between various queues and resource allocation among jobs. This paper reports some results of the performance evaluation study of the Cyber-176 by modeling the scheduler as the heart of the system. The study explores the effects of varying the scheduler parameters in the performance of the machine in a particular installation. The basic theme of the paper is that the selection of parameters in a laboratory or a system test environment may not always result in the best performance in an actual installation. The simulation provides vital information for installation management and tuning the operating system.