Bibliothèque Intel® MPI Library

Using Multiple DAPL* Providers with the Intel® MPI Library

Introduction

If your MPI program sends messages of drastically different sizes (for example, some 16 byte messages, and some 4 megabyte messages), you want optimum performance at all message sizes.  This cannot easily be obtained with a single DAPL* provider.  This is due to latency being a major factor for smaller messages, but bandwidth being more important for larger messages, and providers often making a tradeoff to improve one or the other.  The Intel® MPI Library, as of Version 4.1 Update 1, now supports up to three providers for a single rank of your job.

  • Développeurs
  • Linux*
  • Serveur
  • Avancé
  • Bibliothèque Intel® MPI Library
  • Interface de transmission de messages
  • dapl
  • Informatique en cluster
  • Intel® Many Integrated Core Architecture
  • -mkl=cluster requires Intel(R) Cluster Studio

    Product versions:

    • Intel(R) C++ Composer XE 2013 SP1 for Linux*
    • Intel(R) Fortran Composer XE 2013 SP1 for Linux
    • Intel(R) Cluster Studio XE 2013 SP1

    Problem description:

    If you use -mkl=cluster on your compilation line, but do not have the environment for Intel® Cluster Studio XE 2013 SP1 setup and installed, you will get linker errors like so:

  • Développeurs
  • Professeurs
  • Étudiants
  • Linux*
  • C/C++
  • Fortran
  • Compilateur Intel® C++
  • Intel® C++ Composer XE
  • Compilateur Intel® Fortran
  • Intel® Fortran Composer XE
  • Intel® Visual Fortran Composer XE
  • Bibliothèque Intel® Math Kernel Library
  • Bibliothèque Intel® MPI Library
  • Intel® Cluster Studio XE
  • Best Known Method: Avoid heterogeneous precision in control flow calculations

    Best Known Method

    Running an MPI program in symmetric mode on an Intel® Xeon® host and an Intel Xeon Phi™ coprocessor may deadlock in specific cases due to the heterogeneous precision in replicated control flow calculations. The advice is to determine the control flow only on one master MPI process.

    The Issue

    Intel MPI applications can be executed on multiple combinations of Intel Xeon processors and Intel Xeon Phi coprocessors. Four different execution models are supported:

  • Développeurs
  • Professeurs
  • Étudiants
  • C/C++
  • Fortran
  • Intermédiaire
  • Compilateur Intel® C++
  • Bibliothèque Intel® MPI Library
  • MPI
  • computational accuracy
  • solver control
  • Interface de transmission de messages
  • Intel® Many Integrated Core Architecture
  • Informatique parallèle
  • Intel Software Tools Technical Webinar Series

    These free technical webinars cover tips and techniques that will help sharpen your development skills to create faster, more reliable applications. Technical experts will cover topics ranging from vectorization, code migration, code optimization, using advanced threading techniques (e.g., OpenMP 4.0, Intel® Cilk™ Plus, Intel® TBB), and error checking. Bring programming questions to the live session for our technical experts to answer. A replay of each webinar will be available shortly after the live session so you can share with those unable to attend the live session.

  • Développeurs
  • C/C++
  • Fortran
  • Intel® Cilk™ Plus
  • Bibliothèque Intel® MPI Library
  • Intel® Threading Building Blocks
  • Intel® Cluster Studio XE
  • Intel® Fortran Studio XE
  • Intel® Parallel Studio XE
  • Intel® Advisor XE
  • Amplificateur Intel® VTune™ XE
  • Intel® Inspector XE
  • Webinars
  • Intel® Many Integrated Core Architecture
  • Message Passing Interface (MPI) on Intel® Xeon Phi™ coprocessor

    This two day webinar series introduces you to the world of multicore and manycore computing with Intel® Xeon® processors and Intel® Xeon Phi™ coprocessors. Expert technical teams at Intel discuss development tools, programming models, vectorization, and execution models that will get your development efforts powered up to get the best out of your applications and platforms.

  • Développeurs
  • Serveur
  • Collecteur et analyseur de traces Intel®
  • Bibliothèque Intel® MPI Library
  • Intel Xeon Phi
  • Intel Xeon Phi Coprocessor
  • Interface de transmission de messages
  • Intel® Many Integrated Core Architecture
  • Using Regular Expressions with the Intel® MPI Library Automatic Tuner

    The Intel® MPI Library includes an Automatic Tuner program, called mpitune.  You can use mpitune to find optimal settings for both a cluster and for a specific application.  In order to tune a specific application (or to use a benchmark other than the default for a cluster-specific tuning), mpitune needs a way to determine the performance from a run of the application.  By default, the runtime of the application will be used.  But depending on the application, this could be inaccurate.  You might want only the time for a specific portion of the application, or you might want a different mea

  • Développeurs
  • Serveur
  • Intermédiaire
  • Bibliothèque Intel® MPI Library
  • Automatic Tuning
  • regular expressions
  • Interface de transmission de messages
  • Informatique en cluster
  • Intel® MPI Library 4.1 Update 1 Readme

    The Intel® MPI Library for Linux* and Windows* is a high-performance interconnect-independent multi-fabric library implementation of the industry-standard Message Passing Interface, v2.2 (MPI-2.2) specification. This package is for MPI users who develop on and build for IA-32 and 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.

  • Développeurs
  • Linux*
  • Microsoft Windows* (XP, Vista, 7)
  • Serveur
  • C/C++
  • Fortran
  • Bibliothèque Intel® MPI Library
  • Interface de transmission de messages
  • Informatique en cluster
  • Increase Cluster MPI Application Performance with a "MPI Tune" Up

    The Intel® MPI Library offers many configuration options for improving application performance on your cluster. The number of options can be daunting to a new user. To assist with this, the Automatic Tuning utility is provided. This allows you to tune runtime parameters to match your system and your applications. In this webinar, I will introduce the tuner and provide use cases to begin using it for your cluster and applications.

  • Développeurs
  • Bibliothèque Intel® MPI Library
  • Intel® Cluster Studio XE
  • Interface de transmission de messages
  • OpenMP*
  • HPC
  • High performance computing
  • Informatique en cluster
  • Intel® Many Integrated Core Architecture
  • S’abonner à Bibliothèque Intel® MPI Library