Using Queue Time Predictions for Processor Allocation

Abstract

When a malleable job is submitted to a space-sharing parallel computer, it must choose often whether to begin execution on a small, available cluster, or wait in queue for more processors to become available. To make this decision, it must predict how long it will have to wait for the larger cluster. We propose statistical techniques for predicting these queue times, and develop an allocation strategy that uses these predictions. We present a workload model based on the environment we have observed at the San Diego Supercomputer Center, and use this model to drive simulations of various allocation strategies. We conclude that prediction-based allocation not only improves the average turnaround time for the jobs; it also improves the utilization of the system as a whole.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jan 01, 1997
Accession Number
ADA603824

Entities

People

  • Allen B. Downey

Organizations

  • University of California, Berkeley

Tags

Communities of Interest

  • Energy and Power Technologies
  • Human Systems

DTIC Thesaurus Topics

  • Abstracts
  • Accuracy
  • California
  • Computer Science
  • Computers
  • Control Simulators
  • Design Criteria
  • Environment
  • Measurement
  • Motivation
  • Observation
  • Parallel Computing
  • Scheduling (Production)
  • Simulations
  • Simulators
  • Supercomputers
  • Workload

Fields of Study

  • Computer science

Readers

  • Parallel and Distributed Computing.

Technology Areas

  • Space