Maximize Your Simulation Productivity with ANSYS HPC and NVIDIA GPUs - Presentation
To accurately predict product performance, engineers must conduct a large number of reliable simulations to explore various parameters and collect insight on different design variables (rather than simulate limited scenarios) in a virtual product development process. To keep simulation times within a reasonable range, ANSYS customers take advantage of high-performance computing (HPC) products along with physics-based solvers to run simulation in parallel on multiple-CPU cores. To leverage new and evolving hardware innovations, ANSYS offers GPU acceleration in several of its flagship products, including ANSYS Fluent and ANSYS Mechanical. Used in conjunction with CPUs, GPUs accelerate simulation by handling parallel tasks in matrix operations very efficiently on thousands of cores, reducing simulation times by at least half. This presentation from NVIDIA highlights the typical CFD and structural mechanics applications that can be accelerated with GPUs, including the performance advantages from selected benchmarks and customer cases.