Dev Hub

Optimizing Fluent for Intel Instruction Set Architecture

Abstract

Ansys Fluent (Computational Fluid Dynamics) is one of the most widely used CAE applications in the industry. In this presentation we will cover optimizing Fluent for Intel architecture employing Intel® oneAPI Math Kernel Library (oneMKL) Sparse BLAS kernels. Intel® oneMKL provides substitutions for two of the smoothers most commonly used in Fluent AMG solver. Newly developed oneMKL functionality is based on two steps: an inspection step to analyze different matrix properties, such as sparsity pattern, to facilitate more efficient storage, and execute step that leverages AVX-512 instructions. The work was organized in a way to ease the transition from POC to final product and customer visibility.

Speakers

Maria Zhukova, Software Application Engineer, Intel Corporation

Maria Zhukova is a Software Engineer with seven years of experience working first with Intel Math Kernel Library and then as a Software Enabling and Optimization Engineer focused on HPC. Maria specializes in Computational Mathematics, particularly in Sparse Linear Algebra domain. Maria graduated from the Novosibirsk State University in 2015 with a Master’s Degree in Applied Mathematics and Computer Science, and relocated to Oregon in 2017.

Robert Larson, Software Application Engineer, Intel Corporation

Robert Larson has been working with HPC simulation software at Intel for 16 years. Previously, Robert worked at Cray Research and SGI, based in Minnesota. Robert has a PhD in Chemical Engineering from the University of Illinois in Urbana-Champaign.