首页 | 本学科首页   官方微博 | 高级检索  
     

利用循环分割和循环展开避免Cache代价
引用本文:刘 利,陈 彧,乔 林,汤志忠.利用循环分割和循环展开避免Cache代价[J].软件学报,2008,19(9):2228-2242.
作者姓名:刘 利  陈 彧  乔 林  汤志忠
作者单位:清华大学,计算机科学与技术系,北京,100084
摘    要:存储系统与处理器之间的速度差距逐渐变大,为此,cache使用了分级机制,但这也带来了额外的存储延迟(cache代价).提出一种利用循环分割和循环展开相结合避免cache代价的PCPLPU(prevent cache penalty by loop partition-unrolling)算法.实验结果表明,PCPLPU算法能够有效避免循环代价,提高程序性能.

关 键 词:循环分割  循环展开  cache代价  bank冲突
收稿时间:2005/10/8 0:00:00
修稿时间:2006/7/10 0:00:00

Optimization to Prevent Cache Penalty by Loop Partition and Loop Unrolling
LIU Li,CHEN Yu,QIAO Lin and TANG Zhi-Zhong.Optimization to Prevent Cache Penalty by Loop Partition and Loop Unrolling[J].Journal of Software,2008,19(9):2228-2242.
Authors:LIU Li  CHEN Yu  QIAO Lin and TANG Zhi-Zhong
Abstract:Due to the increasing speed gap between memory system and processor,cache hierarchies have been implemented into memory system,but additional latency(cache penalty)is introduced.This paper presents an algorithm named as prevent cache penalty by loop partition-unrolling(PCPLPU),which can prevent cache penalty in loops by the combination of loop partition and unrolling.Experimental results show that PCPLPU can prevent cache penalty and improve the performance of programs.
Keywords:loop partition  loop unrolling  cache penalty  bank conflict
本文献已被 CNKI 维普 万方数据 等数据库收录!
点击此处可从《软件学报》浏览原始摘要信息
点击此处可从《软件学报》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号