A comparison of search heuristics for empirical code optimization

A comparison of search heuristics for empirical code optimization
复制标题

经验代码优化的搜索启发式比较

DOI:
10.1109/clustr.2008.4663803
复制
发表时间:
2008
期刊:
2008 IEEE International Conference on Cluster Computing
影响因子:
--
通讯作者:
J. Dongarra
J. Dongarra
中科院分区:
--
文献类型:
--
作者:
Keith Seymour;Haihang You;J. Dongarra

文献摘要

被引文献

相似文献

本文描述了各种搜索技术在自动经验代码优化问题中的应用。搜索过程是自动调整系统的一个关键方面,因为搜索空间的大小和评估候选实现的成本使得通过强力找到真正的最佳点是不可行的。我们根据在不同时间限制下找到的最佳候选者的性能来评估 Nelder-Mead 单纯形、遗传算法、模拟退火、粒子群优化、正交搜索和随机搜索的有效性。
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.