Configure mpi

  • mpi-selector --user --set mvapich2_gcc_qlc-1.7
  • export MPI_HOME=/usr/mpi/gcc/mvapich2-1.7-qlc
  • export MPI_INCLUDE=/usr/mpi/gcc/mvapich2-1.7-qlc/include
  • export GCC=/usr/bin/gcc

parallel shell:

pdsh -w lxbrf63c0[2-7],lxbrf65c0[3-7] ''

Latency measurement for 15 nodes. Node number 16 used as a front end machine, for mpirun. It used also to host software RAID1 NFS shared storage, and Infiniband subnet manager.

-bash-4.1$ mpi-selector --query



osu_latency, point to point. 225 measurement points = 15 nodes * 15 nodes. Diagonal of the matrix represent measurement done inside single node MPI over shared memory.

  • techlab_inifiniband_latency_measeurement_openmpi.png:
Edit | Attach | Watch | Print version | History: r1 | Backlinks | Raw View | WYSIWYG | More topic actions
Topic revision: r1 - 2016-05-26 - AritzBrosaIartza
    • Cern Search Icon Cern Search
    • TWiki Search Icon TWiki Search
    • Google Search Icon Google Search

    HardwareLabs/HardwareLabsPublic All webs login

This site is powered by the TWiki collaboration platform Powered by PerlCopyright & 2008-2020 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
or Ideas, requests, problems regarding TWiki? use Discourse or Send feedback