Linger Longer: Fine-Grain Cycle Stealing for Networks of Workstations

Abstract

Studies have shown that a significant fraction of the time, workstations are idle. This paper presents a new scheduling policy called Linger-Longer that exploits the fine-grained availability of workstations to run sequential and parallel jobs. The authors present a two-level workload characterization study and use it to simulate a cluster of work stations running the new policy. They compare two variations of this policy to two previous policies: Immediate-Eviction and Pause-and-Migrate. This study shows that the Linger-Longer policy can improve the throughput of foreign jobs on cluster by 60% with only a 0.5% slow- down of foreground jobs. For parallel computing, we showed that the Linger-Longer policy outpeforms reconfiguration strategies when the processor utilization by the local process is 20% or less in both synthetic bulk synchronous and real data-parallel applications.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
May 01, 1997
Accession Number
ADA376812

Entities

People

  • Jeffrey K. Hollingsworth
  • Kyung D. Ryu

Organizations

  • University of Maryland

Tags

Communities of Interest

  • Materials and Manufacturing Processes

DTIC Thesaurus Topics

  • Computations
  • Computer Science
  • Computers
  • Cost Models
  • Costs
  • Graphs
  • Intervals
  • Migration
  • Models
  • Molecular Dynamics
  • Operating Systems
  • Parallel Computing
  • Scheduling (Production)
  • Simulations
  • Simulators
  • Throughput
  • Work Stations

Fields of Study

  • Computer science

Readers

  • Parallel and Distributed Computing.