A comparison of search heuristics for empirical code optimization

Keith Seymour, Haihang You, Jack Dongarra

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

44 Scopus citations

Abstract

This paper describes the application of various search techniques to the problem of automatic empirical code optimization. The search process is a critical aspect of auto-tuning systems because the large size of the search space and the cost of evaluating the candidate implementations makes it infeasible to find the true optimum point by brute force. We evaluate the effectiveness of Nelder-Mead Simplex, Genetic Algorithms, Simulated Annealing, Particle Swarm Optimization, Orthogonal search, and Random search in terms of the performance of the best candidate found under varying time limits.

Original languageEnglish
Title of host publicationProceedings of the 2008 IEEE International Conference on Cluster Computing, CCGRID 2008
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages421-429
Number of pages9
ISBN (Print)9781424426409
DOIs
StatePublished - 2008
Event2008 IEEE International Conference on Cluster Computing, ICCC 2008 - Tsukuba, Japan
Duration: Sep 29 2008Oct 1 2008

Publication series

NameProceedings - IEEE International Conference on Cluster Computing, ICCC
VolumeProceedings of the 2008 IEEE International Conference on Clus...
ISSN (Print)1552-5244

Conference

Conference2008 IEEE International Conference on Cluster Computing, ICCC 2008
Country/TerritoryJapan
CityTsukuba
Period09/29/0810/1/08

Fingerprint

Dive into the research topics of 'A comparison of search heuristics for empirical code optimization'. Together they form a unique fingerprint.

Cite this