A newer version of this document is available. Customers should click here to go to the newest version.
What’s New
This section provides changes for the Intel(R) MPI Benchmarks as compared to the previous versions of this product.
Intel(R) MPI Benchmarks 2019 Update 7
No changes except bug fixes.
Intel(R) MPI Benchmarks 2019 Update 6
New IMB-P2P Stencil2D and Stencil3D benchmarks.
Intel(R) MPI Benchmarks 2019 Update 5
Added Microsoft Visual Studio(R) projects for IMB-P2P.
Intel(R) MPI Benchmarks 2019 Update 4
No changes except bug fixes.
Intel(R) MPI Benchmarks 2019 Update 3
Added the warm_up option, which enables additional cycles before running benchmark (for all sizes).
Intel(R) MPI Benchmarks 2019 Update 2
New IMB-P2P benchmarks. IMB-P2P is a shared memory transport oriented benchmarks for MPI-1 point-to-point communications. See IMB-P2P Benchmarks for details.
Intel(R) MPI Benchmarks 2019 Update 1
Added new options for MPI-1. See Command-Line Control for details.
Iteration policy can no longer be set with the -iter option. Use -iter_policy instead.
Intel(R) MPI Benchmarks 2019
New IMB-MT benchmarks. The benchmarks implement the multithreaded version of some of the IMB-MPI1 benchmarks using the OpenMP* paradigm. See Multithreaded MPI-1 Benchmarks for details.
Added a new benchmark. See Reduce_scatter_block for details.
Changes in syntax for the -include and -exclude options. Benchmarks to include and exclude now must be separated by a comma rather than a space. Benchmarks to launch can be separated by a comma or a space. See Command-Line Control for details.
Added new options for MPI-1. See Command-Line Control for details.
Iteration policy can no longer be set with the -iter option. Use -iter_policy instead.
Intel(R) MPI Benchmarks 2018 Update 1
Support for the Microsoft* Visual Studio 2017. Microsoft Visual Studio 2012 support is removed.
Intel(R) MPI Benchmarks 2018
Extended description for the -multi and -map options. See Command-Line Control.
Removed support of the Intel(R) Xeon Phi* coprocessor (formerly code named Knights Corner).
Intel(R) MPI Benchmarks 2017 Update 1
This release includes the following updates as compared to the Intel(R) MPI Benchmarks 2017:
Added a new option -imb_barrier.
The PingPong and PingPing benchmarks are now equivalent to PingPongSpecificSource and PingPingSpecificSource, respectively. Their old behavior (with MPI_ANY_SOURCE) is available in PingPongAnySource and PingPingAnySource.
Intel(R) MPI Benchmarks 2017
This release includes the following updates as compared to the Intel(R) MPI Benchmarks 4.1 Update 1:
Changed default values for the -sync and -root_shift options. See Command-Line Control.
Support for the Microsoft Visual Studio* 2015. Microsoft Visual Studio 2010 support is removed.
Minor improvements and bug fixes.
Intel(R) MPI Benchmarks 4.1 Update 1
This release includes the following updates as compared to the Intel(R) MPI Benchmarks 4.1:
Minor improvements and bug fixes.
Intel(R) MPI Benchmarks 4.1
This release includes the following updates as compared to the Intel(R) MPI Benchmarks 4.0:
Introduced new benchmarks: Uniband and Biband. See Parallel Transfer Benchmarks.
Introduced new options: -sync and -root_shift. See Command-Line Control.
Intel(R) MPI Benchmarks 4.0
This release includes the following updates as compared to the Intel(R) MPI Benchmarks 3.2.4:
Introduced new components IMB-NBC and IMB-RMA that conform to the MPI-3.0 standard.
Introduced a new feature to set the appropriate policy for automatic calculation of iterations. You can set the policy using the -iter and -iter_policy options.
Added new targets to the Linux* OS Makefiles:
-
NBC for building IMB-NBC
RMA for building IMB-RMA
Updated Microsoft* Visual Studio* solutions to include the IMB-NBC and IMB-RMA targets.
Support for the Microsoft* Visual Studio* 2013. Microsoft* Visual Studio* 2008 support is removed.