咨询与建议

限定检索结果

文献类型

  • 1 篇 会议

馆藏范围

  • 1 篇 电子文献
  • 0 种 纸本馆藏

日期分布

学科分类号

  • 1 篇 工学
    • 1 篇 计算机科学与技术...
    • 1 篇 软件工程

主题

  • 1 篇 design space
  • 1 篇 runtime api call...
  • 1 篇 indexes
  • 1 篇 gpgpu
  • 1 篇 prefetching time...
  • 1 篇 dynamic data pre...
  • 1 篇 multi-threading
  • 1 篇 thread block ide...
  • 1 篇 cuda
  • 1 篇 arrays
  • 1 篇 accessed memory ...
  • 1 篇 performance impr...
  • 1 篇 api calls
  • 1 篇 kernel parsing
  • 1 篇 thread block
  • 1 篇 memory requests
  • 1 篇 program diagnost...
  • 1 篇 thread block gra...
  • 1 篇 hardware-softwar...
  • 1 篇 software control

机构

  • 1 篇 univ victoria el...

作者

  • 1 篇 baniasadi amiral...
  • 1 篇 lashgar ahmad

语言

  • 1 篇 英文
检索条件"主题词=runtime API calls"
1 条 记 录,以下是1-10 订阅
排序:
Rethinking Prefetching in GPGPUs: Exploiting Unique Opportunities  17
Rethinking Prefetching in GPGPUs: Exploiting Unique Opportun...
收藏 引用
2015 IEEE 17th International Conference on High Performance Computing and Communications (HPCC)
作者: Lashgar, Ahmad Baniasadi, Amirali Univ Victoria Elect & Comp Engn Victoria BC Canada
In this paper we investigate static memory access predictability in GPGPU workloads, at the thread block granularity. We first show that a significant share of accessed memory addresses can be predicted using thread b... 详细信息
来源: 评论