A refactoring method for cache-efficient swarm intelligence algorithms
學年 100
學期 2
出版（發表）日期 2012-06-01
作品名稱 A refactoring method for cache-efficient swarm intelligence algorithms
作品名稱（其他語言）
著者 Chang, Feng-Cheng; Huang, Hsiang-Cheh
單位 淡江大學資訊創新與科技學系
出版者 Philadelphia, PA: Elsevier Inc.
著錄名稱、卷期、頁數 Information Sciences 192, pp.39–49
摘要 With advances in hardware technology, conventional approaches to software development are not effective for developing efficient algorithms for run-time environments. The problem comes from the overly simplified hardware abstraction model in the software development procedure. The mismatch between the hypothetical hardware model and real hardware design should be compensated for in designing an efficient algorithm. In this paper, we focus on two schemes: one is the memory hierarchy, and the other is the algorithm design. Both the cache properties and the cache-aware development are investigated. We then propose a few simple guidelines for revising a developed algorithm in order to increase the utilization of the cache. To verify the effectiveness of the guidelines proposed, optimization techniques, including particle swarm optimization (PSO) and the genetic algorithm (GA), are employed. Simulation results demonstrate that the guidelines are potentially helpful for revising various algorithms.
關鍵字 Cache;Memory hierarchy;Miss rate;Swarm intelligence;Particle swarm optimization;Genetic algorithm
語言 en
ISSN 0020-0255
期刊性質 國外
收錄於 SCI
產學合作
通訊作者 Huang, Hsiang-Cheh
審稿制度 是
國別 USA
公開徵稿
出版型式 紙本
相關連結

機構典藏連結 ( http://tkuir.lib.tku.edu.tw:8080/dspace/handle/987654321/78779 )

機構典藏連結
SDGS 優質教育,產業創新與基礎設施