| 片上多处理器中基于步长和指针的预取 |
| |
| 引用本文： | 肖俊华,冯子军,章隆兵.片上多处理器中基于步长和指针的预取[J].计算机工程,2009,35(4):58-60. |
| |
| 作者姓名： | 肖俊华 冯子军 章隆兵 |
| |
| 作者单位： | 1. 中国科学院计算技术研究所,北京,100080;中国科学院研究生院,北京,100039
2. 中国科学院计算技术研究所,北京,100080 |
| |
| 摘 要： | 在对大量程序访存行为进行分析的基础上，提出基于步长和指针的预取方法。能捕获规整的数据访问模式和指针访问模式。在L2cache和内存之间采用全局历史缓存实现该预取方法。全系统模拟结果表明，该预取方法对商业应用测试程序的性能平均提高14％，对科学计算测试程序的性能平均提高34．5％。
|
| 关 键 词： | 片上多处理器 步长预取 指针预取 |
| 修稿时间： | |
Stride and Pointer Based Prefetching in Chip Multiprocessor |
| |
| XIAO Jun-hua,FENG Zi-jun,ZHANG Long-bing.Stride and Pointer Based Prefetching in Chip Multiprocessor[J].Computer Engineering,2009,35(4):58-60. |
| |
| Authors: | XIAO Jun-hua FENG Zi-jun ZHANG Long-bing |
| |
| Affiliation: | 1.Institue of Computing Technology;Chinese Academy of Sciences;Beijing 100080;2.Graduate University of Chinese Academy of Sciences;Beijing 100039 |
| |
| Abstract: | Based on the analysis of benchmark program’s memory access behavior, this paper proposes stride and pointer based prefetching scheme which can capture regular data access pattern and pointer access pattern, and implements a prefetcher between L2 cache and memory using a global history buffer. Full system simulation shows that this prefetching scheme can improve the commercial benchmark’s performance by 14%, and improve the scientific benchmark’s performance by 34.5%. |
| |
| Keywords: | chip multiprocessor stride prefetching pointer prefetching |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
| 点击此处可从《计算机工程》浏览原始摘要信息 |
|
点击此处可从《计算机工程》下载全文 |
