Running Intel® MPI Library in Containers

A container is a self-contained execution environment platform that enable flexibility and portability of your MPI appliction. It lets you package an application and its dependencies in a virtual container that can run on an operating system, such as Linux*.

This guide describes the use of the Intel® MPI Library with the Singularity* container type.

Singularity Containers

Singularity* is a lightweight container model aligned with the needs of high performance computing (HPC). Singularity has a built-in support of MPI and allows you to leverage the resources of the host you are on, including HPC interconnects, resource managers, and accelerators.

This chapter provides information on running Intel® MPI Library in a Singularity container built from a recipe file. To run Intel® MPI Library in a Singularity environment, do the following:

  1. Make sure you have the following components installed on each machine of a cluster:

    1. Singularity (version not lower than 3.0).
    2. A container including your application.
    3. Intel MPI Library.
  2. Create a Singularity recipe file and use it to build a container.
  3. Run your MPI application from the Singularity container.
Parent topic: Running Applications