首页> 外文期刊>Parallel Computing >Using shared-data localization to reduce the cost of inspector-execution in unified-parallel-C programs
【24h】

Using shared-data localization to reduce the cost of inspector-execution in unified-parallel-C programs

机译:使用共享数据本地化来减少统一并行C程序中检查器执行的成本

获取原文
获取原文并翻译 | 示例

摘要

Programs written in the Unified Parallel C (UPC) language can access any location of the entire local and remote address space via read/write operations. However, UPC programs that contain fine-grained shared accesses can exhibit performance degradation. One solution is to use the inspector-executor technique to coalesce fine-grained shared accesses to larger remote access operations. A straightforward implementation of the inspector executor transformation results in excessive instrumentation that hinders performance.
机译:用Unified Parallel C(UPC)语言编写的程序可以通过读/写操作访问整个本地和远程地址空间的任何位置。但是,包含细粒度共享访问的UPC程序可能会降低性能。一种解决方案是使用检查器-执行器技术来合并对较大的远程访问操作的细粒度共享访问。检查器执行程序转换的直接实现会导致过多的工具,从而妨碍性能。

著录项

  • 来源
    《Parallel Computing》 |2016年第5期|2-14|共13页
  • 作者单位

    Barcelona Supercomp Ctr, C Jordi Girona 1-3, Barcelona 08034, Spain|IBM Canada CAS Res, Warden Ave, Markham, ON L6G 1C7, Canada;

    IBM Toronto Software Lab, Warden Ave, Markham, ON L6G 1C7, Canada;

    Univ Alberta, Dept Comp Sci, Athabasca Hall ATH 342, Edmonton, AB T6G 2E8, Canada;

    Univ Politecn Cataluna, Dept Comp Architecture, C Jordi Girona 1-3, ES-08034 Barcelona, Spain;

    Univ Politecn Cataluna, Dept Comp Architecture, C Jordi Girona 1-3, ES-08034 Barcelona, Spain;

  • 收录信息 美国《科学引文索引》(SCI);美国《工程索引》(EI);
  • 原文格式 PDF
  • 正文语种 eng
  • 中图分类
  • 关键词

    Unified Parallel C; Partitioned global address space; Compiler optimization;

    机译:统一并行C;分区全局地址空间;编译器优化;

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