Using runtime measured workload characteristics in parallel processor scheduling

Thu D. Nguyen, Raj Vaswani, John Zahorjan

Research output: Chapter in Book/Report/Conference proceedingConference contribution

24 Scopus citations

Abstract

We consider the use of runtime measured workload characteristics in parallel processor scheduling. Although many researchers have considered the use of application characteristics in this domain, most of this work has assumed that such information is available a priori. In contrast, we propose and evaluate experimentally dynamic processor allocation policies that rely on determining job characteristics at mntime; in particular, we focus on measuring and using job efficiency and speedup. Our work is intended to be a first step towards the eventual development of production schedulers that use runtime measured workload characteristics in making their decisions. The experimental results we present validate the following observations: - Despite the inherent inaccuracies of runtime measurements and the added overhead of more frequent reallocations, schedulers that use mntime measurements of workload characteristics can significantly outperform schedulers that are oblivious to these characteristics. - Runtime measurements are sufficient for schedulers to achieve performance surprisingly close to that possible when a priori efficiency and speedup information is available. - The primary performance loss, relative to the use of a priori information, is due to the transient decisions of the schedulers as they acquire information on the running applications, rather than to measurement and reallocation overheads. We consider both interactive environments, in which a response time directed scheduler is appropriate, and batch environments, in which maximizing useful instruction throughput is the primary goal. Our experiments are performed using prototype implementations running on a 50-node KSR-2 shared memory multiprocessor.

Original languageEnglish (US)
Title of host publicationJob Scheduling Strategies for Parallel Processing - IPPS 1996 Workshop, Proceedings
EditorsDror G. Feitelson, Larry Rudolph
PublisherSpringer Verlag
Pages155-174
Number of pages20
ISBN (Print)3540618643, 9783540618645
DOIs
StatePublished - 1996
Externally publishedYes
Event2nd Workshop on Job Scheduling Strategies for Parallel Processing, IPPS 1996 - Honolulu, United States
Duration: Apr 16 1996Apr 16 1996

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume1162
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other2nd Workshop on Job Scheduling Strategies for Parallel Processing, IPPS 1996
Country/TerritoryUnited States
CityHonolulu
Period4/16/964/16/96

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • Computer Science(all)

Fingerprint

Dive into the research topics of 'Using runtime measured workload characteristics in parallel processor scheduling'. Together they form a unique fingerprint.

Cite this