Developer Story: How We Ported oneDNN to Fugaku with Arm
Fujitsu optimized and ported the oneDNN DL process library software for the Armv8-A instruction set so that it can be run at high speed on the Fugaku supercomputer. Senior Research for the Fujitsu Laboratories Platform Innovation Project, Kentaro Kawakami, shares how they introduced the JIT assembler, a key technology required for porting oneDNN, as well as the development history which accelerates port development. Fujitsu managed to make full use of Arm SVE architecture, resulting in improving performance 9.2 times in training and 7.8 times in inference. The assembler they developed has been officially incorporated into the main branch of oneDNN along with the source code optimized for the Armv8-A instruction set.
Fujitsu and RIKEN Optimized oneDNN for Improved Performance on ARM
Claiming the title of world’s fastest performance for the number of deep learning models trained per time unit for CosmoFlow, Fujitsu and RIKEN take first...
Powering Heterogenous Computing with oneAPI on ARM and Ponte Vecchio with SiPearl
Intel’s oneDNN 2.1 Released With NVIDIA GPU Support, Initial Alder Lake Optimizations
Out today is a new release of Intel’s open-source oneDNN library used as a deep neural network library for assembling deep learning applications.
