Issue link: https://resources.randsim.com/i/1504493
156 ©2023 ANSYS, Inc. GPU Acceleration Enhancements • Significantly faster performance (Engine block model) 0 1 2 3 4 5 6 7 8 9 10 0 GPU 1 GPU Relative Speedup DMP Performance w/ 4 cores 2023 R2 Intel Xeon Gold 6342 processors, 1 TB RAM, NVIDIA H100 PCIe card, RHEL 8.7 • 4.2 million DOF; sparse solver • Nonlinear static analysis involving contact, plasticity and gasket elements Measuring the matrix factorization algorithm that is mostly offloaded to the GPU