The High Performance LINPACK (HPL) benchmark solves a dense linear system in double precision (64-bit) arithmetic on distributed-memory computers. HPL is written in a portable ANSI C and requires an MPI implementation and either BLAS or the Vector Signal and Image Processing Library (VSIPL). HPL is often one of the first programs to run on large HPC installations, producing a result that can be submitted to the TOP500 list of the world’s fastest supercomputers.
The major focus of HPL 2.2, released in 2016, was to improve the accuracy of reported benchmark results and ensure scalability of the code on large supercomputer installations with 100,000+ cores. HPL now features detailed time-of-run accounting to better assess power requirements at run time, a metric which is reported on the TOP500 and also highlighted on the Green500. The LINPACK app for iOS achieves over 4 gigaFLOP/s on the iPad Air and iPad Pro. For the November 2017 TOP500 list, an optimized version of the HPL code ran on a machine with over 10,000,000 cores.
Find out more at http://icl.utk.edu/hpl/
In Collaboration With
- Oak Ridge National Laboratory