mpi-3

借助 MPI-3 非阻塞集合改进性能

全新 MPI-3 非阻塞集合能够提高应用性能。该改进效果在适当的应用上将会非常显著。但是对于一些应用,如果添加非阻塞集合,将会降低性能。接下来,我将介绍什么是非阻塞集合,并展示一个能够通过使用 MPI_Iallreduce 获益的内核。

什么是 MPI-3 非阻塞集合?

非阻塞集合是能够立即返回到应用代码的新版本集合功能。这些版本可以在您的应用执行其他任务时在后台执行集合操作(只要您的 MPI 实现支持)。如果您的应用结构支持您开始集合操作、执行本地工作并随后从集合操作中获取结果,那么您的应用可以通过使用非阻塞集合获益。

何时使用非阻塞集合?

为了从非阻塞集合获得优势,您的应用在集合开始到必须完成时必须能够执行大量任务,以抵消检查集合完成所额外付出的开销。一般而言,尺寸较大的消息需要更多计算能力来抵消将数据迁移至通信缓冲。如果您与重叠计算相关的消息尺寸较小,则可以获益。

此外,您的可用系统资源必须充足。如果您已经使用了所有可用系统资源,那么 MPI 实现无法与计算并行运行通信,从而将无法提供优势,可能还会降低性能。

  • Sviluppatori
  • Linux*
  • Microsoft Windows* (XP, Vista, 7)
  • Microsoft Windows* 8
  • Server
  • Intermedio
  • Intel® Trace Analyzer and Collector
  • Intel® MPI Library
  • Message Passing Interface
  • mpi-3
  • non-blocking collectives
  • Elaborazione basata su cluster
  • Ottimizzazione
  • Elaborazione parallela
  • Improving Performance with MPI-3 Non-Blocking Collectives

    The new MPI-3 non-blocking collectives offer potential improvements to application performance.  These gains can be significant for the right application.  But for some applications, you could end up lowering your performance by adding non-blocking collectives.  I'm going to discuss what the non-blocking collectives are and show a kernel which can benefit from using MPI_Iallreduce.

  • Sviluppatori
  • Linux*
  • Microsoft Windows* (XP, Vista, 7)
  • Microsoft Windows* 8
  • Server
  • Intermedio
  • Intel® Trace Analyzer and Collector
  • Intel® MPI Library
  • Message Passing Interface
  • mpi-3
  • non-blocking collectives
  • Elaborazione basata su cluster
  • Ottimizzazione
  • Elaborazione parallela
  • Intel® Trace Analyzer and Collector 9.0 Beta Readme

    The Intel® Trace Analyzer and Collector is a low-overhead scalable event-tracing library with graphical analysis that reduces the time it takes an application developer to enable maximum performance of cluster applications. This Beta package is for users who develop on and build for Intel® 64 architectures on Linux* and Windows*, as well as customers running on the Intel® Xeon Phi™ coprocessor on Linux*. You must have a valid license to download, install and use this product.

  • Linux*
  • Microsoft Windows* (XP, Vista, 7)
  • Microsoft Windows* 8
  • C/C++
  • Fortran
  • Intel® Trace Analyzer and Collector
  • mpi-3
  • mpi analysis
  • profiling
  • cluster tools
  • Message Passing Interface
  • Elaborazione basata su cluster
  • Intel® MPI Library 5.0 Beta Readme

    The Intel® MPI Library is a high-performance interconnect-independent multi-fabric library implementation of the industry-standard Message Passing Interface, v3.0 (MPI-3.0) specification. This Beta package is for MPI users who develop on and build for Intel® 64 architectures on Linux* and Windows*, as well as customers running on the Intel® Xeon Phi™ coprocessor on Linux*. You must have a valid license to download, install and use this product.

  • Linux*
  • Microsoft Windows* (XP, Vista, 7)
  • C/C++
  • Intel® MPI Library
  • MPI
  • mpi-3
  • cluster studio
  • Message Passing Interface
  • Elaborazione basata su cluster
  • Iscriversi a mpi-3