Developer Guide

Developer Guide for Intel® oneAPI Math Kernel Library Linux*

ID 766690
Date 7/13/2023
Public

A newer version of this document is available.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

Document Table of Contents

Contents of the Intel® Distribution for LINPACK* Benchmark and the Intel® Optimized HPL-AI* Benchmark

The Intel® Distribution for LINPACK Benchmark and the Intel® Optimized HPL-AI Benchmark include prebuilt binaries linked with Intel® MPI library. For a customized MPI implementation, tools are also included to build a binary using Intel® oneAPI Math Kernel Library (oneMKL) MPI wrappers. All the files are located in the./benchmarks/mp_linpack/subdirectory of the Intel® oneAPI Math Kernel Library (oneMKL) directory.

File in <mkl directory>/benchmarks/mp_linpack/

Contents

COPYRIGHT

Original Netlib HPL copyright document

readme.txt

Information about the files provided

Prebuilt executables for performance testing

xhpl_intel64_dynamic

Prebuilt binary of the Intel® Distribution for LINPACK Benchmark for CPU

xhpl_intel64_dynamic_gpu

Prebuilt binary of the Intel® Distribution for LINPACK Benchmark for GPU

xhpl-ai_intel64_dynamic

Prebuilt binary of the Intel® Optimized HPL-AI Benchmark for CPU

xhpl-ai_intel64_dynamic_gpu

Prebuilt binary of the Intel® Optimized HPL-AI Benchmark for GPU

Run scripts and an input file example

runme_intel64_dynamic

Sample run script for the Intel® 64 architecture and binary dynamically linked against Intel MPI library

runme_intel64_prv

Script that sets HPL environment variables. It is called by runme_intel64_dynamic

HPL.dat

Example of an HPL configuration file

Prebuilt libraries and utilities for building with a customized MPI implementation

libhpl_intel64.a, libhpl_inte64_gpu.a, libhpl-ai_intel64.a, libhpl-ai_intel64_gpu.a

Library file required to build Intel® Distribution for LINPACK Benchmark for the Intel® 64 architecture with a customized MPI implementation

HPL_main.c

Source code required to build Intel® Distribution for LINPACK Benchmark for the Intel® 64 architecture with a customized MPI implementation

build.sh

Build script for creating Intel® Distribution for LINPACK Benchmark for the Intel® 64 architecture with a customized MPI implementation

Product and Performance Information

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

Notice revision #20201201