Accurate modeling of parallel scientific computations

David M. Nicol, James C. Townsend

Research output: Contribution to journalConference articlepeer-review

Abstract

Scientific codes are usually parallelized by partitioning a grid among processors. To achieve top performance it is necessary to partition the grid so as to balance workload and minimize communication/synchronization costs. This problem is particularly acute when the grid is irregular, changes over the course of the computation, and is not known until load-time. Critical mapping and remapping decisions rest on our ability to accurately predict performance, given a description of a grid and its partition. This paper discusses one approach to this problem, and illustrates its use on a one-dimensional fluids code. The models we construct are shown empirically to be accurate, and are used to find optimal remapping schedules.

Original languageEnglish (US)
Pages (from-to)165-170
Number of pages6
JournalPerformance Evaluation Review
Volume17
Issue number1
DOIs
StatePublished - 1989
Externally publishedYes
EventACM Sigmetrics and Performance '89 International Conference on Measurement and Modeling of Computer Systems - Proceedings - Berkeley, CA, USA
Duration: May 23 1989May 26 1989

ASJC Scopus subject areas

  • Software
  • Hardware and Architecture
  • Computer Networks and Communications

Fingerprint

Dive into the research topics of 'Accurate modeling of parallel scientific computations'. Together they form a unique fingerprint.

Cite this