Institute of Computing Technology, Chinese Academy IR
Reinventing Memory System Design for Many-Accelerator Architecture | |
Wang Ying1; Zhang Lei1; Han YinHe1; Li HuaWei1 | |
2014 | |
发表期刊 | JOURNAL OF COMPUTER SCIENCE AND TECHNOLOGY |
ISSN | 1000-9000 |
卷号 | 29期号:2页码:273 |
摘要 | The many-accelerator architecture, mostly composed of general-purpose cores and accelerator-like function units (FUs), becomes a great alternative to homogeneous chip multiprocessors (CMPs) for its superior power-efficiency. However, the emerging many-accelerator processor shows a much more complicated memory accessing pattern than general purpose processors (GPPs) because the abundant on-chip FUs tend to generate highly-concurrent memory streams with distinct locality and bandwidth demand. The disordered memory streams issued by diverse accelerators exhibit a mutual-interference behavior and cannot be efficiently handled by the orthodox main memory interface that provides an inflexible data fetching mode. Unlike the traditional DRAM memory, our proposed Aggregation Memory System (AMS) can function adaptively to the characterized memory streams from different FUs, because it provides the FUs with different data fetching sizes and protects their locality in memory access by intelligently interleaving their data to memory devices through sub-rank binding. Moreover, AMS can batch the requests without sub-rank conflict into a read burst with our optimized memory scheduling policy. Experimental results from trace-based simulation show both conspicuous performance boost and energy saving brought by AMS. |
关键词 | EFFICIENCY CACHE many-accelerator chip multiprocessor memory general purpose processor |
语种 | 英语 |
资助项目 | [National Natural Science Foundation of China] ; [National Basic Research 973 Program of China] ; [Strategic Priority Research Program of the Chinese Academy of Sciences] |
文献类型 | 期刊论文 |
条目标识符 | http://119.78.100.204/handle/2XEOYT63/35516 |
专题 | 中国科学院计算技术研究所期刊论文_中文 |
作者单位 | 1.中国科学院计算技术研究所 2.中国科学院大学 |
第一作者单位 | 中国科学院计算技术研究所 |
推荐引用方式 GB/T 7714 | Wang Ying,Zhang Lei,Han YinHe,et al. Reinventing Memory System Design for Many-Accelerator Architecture[J]. JOURNAL OF COMPUTER SCIENCE AND TECHNOLOGY,2014,29(2):273. |
APA | Wang Ying,Zhang Lei,Han YinHe,&Li HuaWei.(2014).Reinventing Memory System Design for Many-Accelerator Architecture.JOURNAL OF COMPUTER SCIENCE AND TECHNOLOGY,29(2),273. |
MLA | Wang Ying,et al."Reinventing Memory System Design for Many-Accelerator Architecture".JOURNAL OF COMPUTER SCIENCE AND TECHNOLOGY 29.2(2014):273. |
条目包含的文件 | 条目无相关文件。 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论