Processor preallocation and load balancing of DOALL loops

Gary W. Elsesser, Viet N. Ngo, Sourav Bhattacharya, Wei Tek Tsai

Research output: Contribution to journalArticlepeer-review

1 Scopus citations

Abstract

Load balance is important because it may affect the speedup attained through the concurrent execution of loop iterations on a parallel processor. We study loop load balance in the context of the well-known Perfect benchmarks. Several static and dynamic characteristics of the Perfect benchmark DOALL loops are observed and interpreted. The late arrival of processors is noted as a major source of load imbalance. This observation suggested the idea of processor preallocation. An analytic cost model is presented and the advantages of processor preallocation are demonstrated by experimental evaluation on a CRAY Y-MP8 under the Unicos operating system.

Original languageEnglish (US)
Pages (from-to)135-161
Number of pages27
JournalThe Journal of Supercomputing
Volume8
Issue number2
DOIs
StatePublished - Jun 1 1994

Keywords

  • DOALL
  • Perfect Club
  • SPMD paradigm
  • loop scheduling
  • shared memory

ASJC Scopus subject areas

  • Software
  • Theoretical Computer Science
  • Information Systems
  • Hardware and Architecture

Fingerprint Dive into the research topics of 'Processor preallocation and load balancing of DOALL loops'. Together they form a unique fingerprint.

Cite this