首页> 外文期刊>Concurrency, practice and experience >Communication-aware thread mapping using the translation lookaside buffer
【24h】

Communication-aware thread mapping using the translation lookaside buffer

机译:使用转换后备缓冲区的通信感知线程映射

获取原文
获取原文并翻译 | 示例

摘要

Threads of parallel applications need to communicate in order to fulfill their tasks. The communicationrnperformance between the cores in modern multi-core architectures differs because of the memory and interconnectionrnhierarchies. In these architectures, it is important to map the threads of parallel applications byrntaking into account the communication between them, to improve their performance and energy consumption.rnIn parallel applications based on shared memory, communication is implicit, which makes it difficultrnto detect the communication pattern between the threads.rnIn this paper, we introduce a new lightweight mechanism to detect the communication pattern betweenrnthreads of shared memory applications using the translation lookaside buffer. Our mechanism relies on hardwarernfeatures, which make it transparent to the programmer and allow the detection to be performed by thernoperating system during the execution of the application. We also developed a heuristic mapping algorithmrnthat uses the detected pattern to dynamically map the threads to cores. Experiments were performed withrnapplications from the NAS-OMP and PARSEC parallel benchmark suites in a simulated machine as well as arnreal machine. Results show that our mechanism can substantially improve parallel application performance,rnas well as processor and DRAM energy consumption.
机译:并行应用程序的线程需要进行通信才能完成其任务。现代多核体系结构中的内核之间的通信性能因内存和互连层次结构而异。在这些体系结构中,重要的是通过考虑并行应用程序之间的通信来映射它们,以提高它们的性能和能耗。在基于共享内存的并行应用程序中,通信是隐式的,这使得检测通信模式变得困难。在本文中,我们引入了一种新的轻量级机制,该机制使用转换后备缓冲区来检测共享内存应用程序的线程之间的通信模式。我们的机制依赖于硬件功能,这使其对程序员透明,并允许在应用程序执行期间由热操作系统进行检测。我们还开发了一种启发式映射算法,该算法使用检测到的模式将线程动态映射到核心。使用来自NAS-OMP和PARSEC并行基准测试套件的应用程序在模拟机器和arnreal机器中进行了实验。结果表明,我们的机制可以显着提高并行应用程序的性能,处理器和DRAM的能耗。

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