Running the Intel® Distribution for LINPACK* Benchmark
To run the Intel® Distribution for LINPACK Benchmark on multiple nodes or on one node with multiple MPI processes, you need to use MPI and either modify
HPL.dat
or use
Ease-of-use Command-line Parameters. The following example describes how to run the dynamically-linked prebuilt Intel® Distribution for LINPACK Benchmark binary using the script provided. To run other binaries, adjust the steps accordingly.
- Load the necessary environment variables for the Intel MPI Library and Intel® compiler:<compiler directory>\env\vars.bat<mpi directory>\env\vars.bat
- InHPL.dat, set the problem sizeNto 10000. Because this setting is for a test run, the problem size should be small.
- For better performance, enable non-uniform memory access (NUMA) on your system and configure to run an MPI process for each NUMA socket as explained below.High-bandwidth Multi-Channel Dynamic Random Access Memory (MCDRAM) on the second-generation Intel® Xeon® Phi processors may appear to be a NUMA node. However, because there are no CPUs on this node, do not run an MPI process for it.
- Refer to your BIOS settings to enable NUMA on your system.
- Set the following variables at the top of therunme_intel64_dynamic.batscript according to your cluster configuration:
- MPI_PROC_NUM
- The total number of MPI processes.
- MPI_PER_NODE
- The number of MPI processes per each cluster node.
- In theHPL.datfile, set the parametersPsandQsso thatPs*Qsequals the number of MPI processes. For example, for 2 processes, setPsto 1 andQsto 2. Alternatively, leave theHPL.datfile as is and launch with-pand-qcommand-line parameters.
- Executerunme_intel64_dynamic.batscript:runme_intel64_dynamic.bat
- Rerun the test increasing the size of the problem until the matrix size uses about 80% of the available memory. To do this, either modifyNsin line 6 ofHPL.dator use the-ncommand-line parameter:
- For 16 GB:40000 Ns
- For 32 GB:56000 Ns
- For 64 GB:83000 Ns
Product and Performance Information
|
---|
Performance varies by use, configuration and other factors. Learn more at
www.Intel.com/PerformanceIndex.
Notice revision #20201201
|