MPI

Putting Your Data and Code in Order: Optimization and Memory – Part 1

This series of two articles discusses how data and memory layout affect performance and suggests specific steps to improve software performance. The basic steps shown in these two articles can yield significant performance gains. These two articles are designed at an intermediate level. It is assumed the reader desires to optimize software performance using common C, C++ and Fortran* programming options.
  • Developers
  • Professors
  • Students
  • C/C++
  • Beginner
  • Intermediate
  • Intel® Math Kernel Library
  • MPI
  • Intel® Advanced Vector Extensions
  • Code Modernization
  • Parallel Computing
  • Vectorization
  • Inscreva-se agora: Workshop em Otimização de Código C/C++ - 28-29/Janeiro

    Participe do Workshop sobre otimização de software (com foco em C/C++) e computação paralela nos dias 28 e 29 de Janeiro no NCC/UNESP para processadores e co-processadores Intel. Data: 28, 29 de Janeiro, 2016 Local: UNESP/NCC - Rua Dr. Bento Teobaldo Ferraz, 271 - Bldg II São Paulo, SP - Brazil 01140-070
  • Developers
  • Partners
  • Professional
  • Professors
  • Students
  • Linux*
  • Server
  • C/C++
  • Beginner
  • Intermediate
  • Intel® Parallel Studio XE
  • HPC
  • C++
  • AVX
  • vetorização
  • computação paralela
  • Multithreading
  • openmp
  • otimização de software
  • MPI
  • Xeon Phi
  • Haswell
  • Cluster
  • Intel® Advanced Vector Extensions
  • Intel® Streaming SIMD Extensions
  • OpenMP*
  • Academic
  • Cluster Computing
  • Code Modernization
  • Development Tools
  • Optimization
  • Parallel Computing
  • Vectorization
  • Reducing Initialization Times of the Intel MPI® Library

    Running large scale Intel® MPI applications on InfiniBand* clusters, one might have recognized an increasing time spend within the MPI_Init() routine. The reason for this behavior are some MPI runtime infrastructure management operations that are necessary in order to make sure that all MPI ranks have a common environment. Having large MPI runs with multiple thousands of ranks, these operations can consume a huge part of the MPI initialization phase time.

    Subscribe to MPI