This section describes the Intel® MPI Library environment variables related to the support of the Intel® Xeon Phi™ coprocessor (code named Knights Landing) based on the Intel® Many Integrated Core Architecture (Intel® MIC Architecture).
No license (express or implied, by estoppel or otherwise) to any intellectual property rights is granted by this document.
Intel disclaims all express and implied warranties, including without limitation, the implied warranties of merchantability, fitness for a particular purpose, and non-infringement, as well as any warranty arising from course of performance, course of dealing, or usage in trade.
This section describes the local options of the Intel® MPI Library's Hydra process manager. Local options are applied only to the argument set they are specified in. Argument sets are separated by a colon ':'.
-n <# of processes> or -np <# of processes>
Use this option to set the number of MPI processes to run with the current argument set.
The Intel® MPI Library Developer Guide explains how to use the Intel® MPI Library in some common usage scenarios. It provides information regarding compiling, running, debugging, tuning and analyzing MPI applications, as well as troubleshooting information.
Intel® MPI Library provides support for the heterogeneous Windows*-Linux* environment. This means that you can run MPI programs on nodes that operate on Windows and Linux OS as single MPI jobs, using the Hydra process manager.
To run a mixed Linux-Windows MPI job, do the following:
Make sure the Intel MPI Library is installed and operable, and the product versions match on all nodes.
On the Windows hosts, make sure the Hydra service is running:
The cluster specific mode is intended for tuning Intel® MPI Library for a specific cluster. For finding optimal settings, mpitune uses a benchmarking program: it runs tests several times with different parameters and searches for the best ones. By default, Intel® MPI Benchmarks are used.
To tune Intel MPI Library in the cluster specific mode, do the following:
Placement of MPI processes over the cluster nodes plays a significant role in application performance. Intel® MPI Library provides several options to control process placement.
By default, when you run an MPI program, the process manager launches all MPI processes specified with -n on the current node. If you use a job scheduler, processes are assigned according to the information received from the scheduler.