A comparison of search heuristics for empirical code optimization

Keith Seymour; Haihang You; Jack Dongarra

doi:10.1109/CLUSTR.2008.4663803

A comparison of search heuristics for empirical code optimization

Keith Seymour, Haihang You, Jack Dongarra

Research output: Chapter in Book/Conference proceeding › Conference contribution

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. © 2008 IEEE.

Original language	English
Title of host publication	Proceedings - IEEE International Conference on Cluster Computing, ICCC\|Proc. IEEE Int. Conf. Cluster Comput. ICCC
Publisher	IEEE
Pages	421-429
Number of pages	8
ISBN (Print)	9781424426409
DOIs	https://doi.org/10.1109/CLUSTR.2008.4663803
Publication status	Published - 2008
Event	2008 IEEE International Conference on Cluster Computing, CCGRID 2008 - Tsukuba Duration: 1 Jul 2008 → …

Conference

Conference	2008 IEEE International Conference on Cluster Computing, CCGRID 2008
City	Tsukuba
Period	1/07/08 → …

Access to Document

10.1109/CLUSTR.2008.4663803

Cite this

@inproceedings{5642af18ba6844658a3bfd7f576d1f26,

title = "A comparison of search heuristics for empirical code optimization",

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. {\textcopyright} 2008 IEEE.",

author = "Keith Seymour and Haihang You and Jack Dongarra",

year = "2008",

doi = "10.1109/CLUSTR.2008.4663803",

language = "English",

isbn = "9781424426409",

pages = "421--429",

booktitle = "Proceedings - IEEE International Conference on Cluster Computing, ICCC|Proc. IEEE Int. Conf. Cluster Comput. ICCC",

publisher = "IEEE",

address = "United States",

note = "2008 IEEE International Conference on Cluster Computing, CCGRID 2008 ; Conference date: 01-07-2008",

}

Seymour, K, You, H & Dongarra, J 2008, A comparison of search heuristics for empirical code optimization. in Proceedings - IEEE International Conference on Cluster Computing, ICCC|Proc. IEEE Int. Conf. Cluster Comput. ICCC. IEEE, pp. 421-429, 2008 IEEE International Conference on Cluster Computing, CCGRID 2008, Tsukuba, 1/07/08. https://doi.org/10.1109/CLUSTR.2008.4663803

TY - GEN

T1 - A comparison of search heuristics for empirical code optimization

AU - Seymour, Keith

AU - You, Haihang

AU - Dongarra, Jack

PY - 2008

Y1 - 2008

N2 - 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. © 2008 IEEE.

AB - 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. © 2008 IEEE.

U2 - 10.1109/CLUSTR.2008.4663803

DO - 10.1109/CLUSTR.2008.4663803

M3 - Conference contribution

SN - 9781424426409

SP - 421

EP - 429

BT - Proceedings - IEEE International Conference on Cluster Computing, ICCC|Proc. IEEE Int. Conf. Cluster Comput. ICCC

PB - IEEE

T2 - 2008 IEEE International Conference on Cluster Computing, CCGRID 2008

Y2 - 1 July 2008

ER -

A comparison of search heuristics for empirical code optimization

Abstract

Conference

Access to Document

Fingerprint

Cite this