本文已被：浏览 1920次 下载 2690次
Received:March 04, 2009
Received:March 04, 2009
中文摘要: 一般的数据处理程序中，计算时间在其中往往只起次要作用，因此访存方式是否有效对程序的性能影响很大。在基于龙芯2F 处理器研制的高性能计算机系统KD-50-I 上安装ATLAS，经测试其性能只达到龙芯2F 理论峰值的30%。通过循环展开减少函数存储访问次数，增大计算访存比；采用数据分块、部分拷贝以增强访存局部性，减少cache 失效；利用非阻塞cache 加快内存访问速度等访存优化技术，将ATLAS 性能提高50%以上。
Abstract:In most cases, compared to computing time, memory access time takes a much larger proportion of program running time. Therefore, memory access approach can affect the program performance significantly. Testing results show that the performance of ATLAS transplanted on KD-50-I, which is based on Loongson 2F,reaches only 30% of its theoretical peak. In this paper, by exploiting Loop Unrolling technique to decrease memory access frequency, enhancing time and space locality to reduce cache misses and nonblocking cache mechanism to form memory access pipeline, the performance of optimized ATLAS can be improved to 50% higher.
keywords: ATLAS KD-50-I cache miss non-blocking cache
文章编号： 中图分类号： 文献标志码：
基金项目:国家高技术研究发展计划(863)(2008AA010902)
| Author Name | Affiliation |
| SU Bo | 中国科学技术大学 计算机科学与技术系 安徽 合肥 230027 |
| LI Kai | |
| XU Zhi-Guang | |
| HE Song-Song |
| Author Name | Affiliation |
| SU Bo | 中国科学技术大学 计算机科学与技术系 安徽 合肥 230027 |
| LI Kai | |
| XU Zhi-Guang | |
| HE Song-Song |
引用文本：
苏波,李凯,徐志广,何颂颂.龙芯2F上的访存优化.计算机系统应用,2010,19(1):171-175
SU Bo,LI Kai,XU Zhi-Guang,HE Song-Song.Optimization of Memory Access Based on Loongson2F.COMPUTER SYSTEMS APPLICATIONS,2010,19(1):171-175
苏波,李凯,徐志广,何颂颂.龙芯2F上的访存优化.计算机系统应用,2010,19(1):171-175
SU Bo,LI Kai,XU Zhi-Guang,HE Song-Song.Optimization of Memory Access Based on Loongson2F.COMPUTER SYSTEMS APPLICATIONS,2010,19(1):171-175
