Developer Guide

Developer Guide for Intel® oneAPI Math Kernel Library Windows*

ID 766692
Date 3/31/2023
Public

A newer version of this document is available. Customers should click here to go to the newest version.

Document Table of Contents

Linking with oneMKL Cluster Software

The Intel® oneAPI Math Kernel Library ScaLAPACK, Cluster FFT, and Cluster Sparse Solver support MPI implementations identified in theIntel® oneAPI Math Kernel Library Release Notes.

To link with ScaLAPACK, Cluster FFT, and/or Cluster Sparse Solver, use the following commands:

set lib =<path to MKL libraries>;<path to MPI libraries>;%lib%

<linker> <files to link> [<MKL cluster library>] <BLACS><MKL core libraries><MPI libraries>

where the placeholders stand for paths and libraries as explained in the following table:

<path to MKL libraries>

<mkl directory>\lib\intel64. If you performed the Scripts to Set Environment Variables Setting Environment Variables step of the Getting Started process, you do not need to add this directory to the lib environment variable.

<path to MPI libraries>

Typically the lib subdirectory in the MPI installation directory.

<linker>

One of icl, ifort, xilink.

<MKL cluster library>

One of libraries for ScaLAPACK or Cluster FFT listed in Appendix C: Directory Structure in Detail. For example, for the LP64 interface, it is mkl_scalapack_lp64.lib or mkl_cdft_core.lib. Cluster Sparse Solver does not require an additional computation library.

<BLACS>

The BLACS library corresponding to your , programming interface (LP64 or ILP64), and MPI version. These libraries are listed in Appendix C: Directory Structure in Detail. For example, for the LP64 interface, choose one of mkl_blacs_intelmpi_lp64.lib, or mkl_blacs_msmpi_lp64.lib in the case of static linking and mkl_blacs_lp64_dll.lib in the case of dynamic linking.

<MKL core libraries>

Intel® oneAPI Math Kernel Library libraries other than libraries with ScaLAPACK, Cluster FFT, or Cluster Sparse Solver.

TIP:

Use the Using the Link-line Advisor to quickly choose the appropriate set of <MKL cluster Library>, <BLACS>, and <MKL core libraries>.

Intel MPI provides prepackaged scripts for its linkers to help you link using the respective linker. Therefore, if you are using Intel MPI, the best way to link is to use the following commands:

<path to Intel MPI binaries>\vars.bat

set lib = <path to MKL libraries>;%lib%

<mpilinker><files to link> [<MKL cluster Library>] <BLACS><MKL core libraries>

where the placeholders that are not yet defined are explained in the following table:

<path to MPI binaries>

By default, the bin subdirectory in the MPI installation directory.

<MPI linker>

mpicl or mpiifort

Product and Performance Information

Performance varies by use, configuration and other factors. Learn more at www.Intel.com/PerformanceIndex.

Notice revision #20201201