OpenBLAS 0.3.6 Boosts AMD Zen Performance For GEMM, Other CPU Optimizations

Written by Michael Larabel in Free Software on 29 April 2019 at 03:24 PM EDT. 3 Comments
FREE SOFTWARE
A new release of OpenBLAS is now available, the widely-used open-source BLAS implementation for optimized linear algebra kernels. With OpenBLAS 0.3.6, work has continued on maximizing the BLAS performance for a variety of CPU architectures.

OpenBLAS 0.3.6 is shipping with CMake build system improvements, various POWER optimizations, support for the HiSilicon TSV110 processors, disabling of the AVX-512 DGEMM kernel once again due to unsolved problems, auto-detection for Intel Denverton CPUs, auto-detection for Hygon Dhyana Zen CPUs, and improved GEMM performance on AMD Zen processors.

OpenBLAS 0.3.6 also carries a variety of fixes, IBM z13/z14 optimizations, and other improvements to this linear algebra library.

More details on the OpenBLAS 0.3.6 changes via the GitHub repository.
Related News
About The Author
Michael Larabel

Michael Larabel is the principal author of Phoronix.com and founded the site in 2004 with a focus on enriching the Linux hardware experience. Michael has written more than 20,000 articles covering the state of Linux hardware support, Linux performance, graphics drivers, and other topics. Michael is also the lead developer of the Phoronix Test Suite, Phoromatic, and OpenBenchmarking.org automated benchmarking software. He can be followed via Twitter, LinkedIn, or contacted via MichaelLarabel.com.

Popular News This Week