Microsoft Windows* 8.x

Interoperability with OpenMP API


The Intel® MPI Library provides an additional environment variable to control process pinning for hybrid MPI/OpenMP* applications. This environment variable is used to define a number of non-overlapping subsets (domains) of logical processors on a node, and a set of rules on how MPI processes are bound to these domains by the following formula: one MPI process per one domain. See the picture.

Fabrics Control

This topic provides you with the information on how to use environment variables to control the following fabrics:

  • Communication fabrics

  • Shared memory fabrics

  • DAPL-capable network fabrics

  • TCP-capable network fabrics

Collective Operation Control

Each collective operation in the Intel® MPI Library supports a number of communication algorithms. In addition to highly optimized default settings, the library provides a way to control the algorithm selection explicitly: I_MPI_ADJUST environment variable family, which is described in the following section.

Subscribe to Microsoft Windows* 8.x