首页> 外文会议>World computer congress >Structure-Based Automatic Extraction of the Program Heterogeneity
【24h】

Structure-Based Automatic Extraction of the Program Heterogeneity

机译:基于结构的自动提取程序异质性

获取原文

摘要

Heterogeneous computing is a new branch of parallel processing and expected to reach super speedup. It is an important step for heterogeneous computing to extract the heterogeneity hiding in a program. Such work is rather difficult and some concept and terminology are ambiguous. From the point of view of program structure, this paper presents the formal definitions about SISD, SIMD, and MIMD, which are the foundations for extracting program heterogeneity. Moreover, some heterogeneity criterions for several types of loop programs are proposed.
机译:异构计算是一个并行处理的新分支,预计将达到超速加速。异构计算是提取在程序中隐藏的异质性的重要步骤。这样的工作相当困难,一些概念和术语是暧昧的。从程序结构的角度来看,本文介绍了关于SISD,SIMD和MIMD的正式定义,这些定义是提取程序异质性的基础。此外,提出了一些用于若干类型环路程序的异构性标准。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号