ANSYS

Computer-aided engineering (CAE) simulations enable engineers to design more virtual prototypes thereby spending less time building physical ones. More iterations leads to a higher quality finished product. GPUs further improve efficiency by accelerating simulation-time.

ANSYS and NVIDIA work closely to ensure the fastest performance for simulations running on ANSYS parallel software. NVIDIA GPUs support several products:

• Mechanical: ANSYS Mechanical 14.5 supports a full feature set including use on multiple GPUs.
• Fluids: ANSYS Fluent 14.5 offers a radiation heat transfer model on GPUs, along with a beta feature of an AMG solver on single GPU.
• Electrical: ANSYS Nexxim 14.5 supports a full feature set.

ANSYS Mechanical 14.5 provides significant performance speedups when using the latest NVIDIA Tesla K20 GPU. See the performance results below.
ANSYS Fluent: The initial implementation of ANSYS Fluent 14.5 is designed to run on a single GPU, and in the future on a single node with multiple GPUs, or across multiple nodes with multiple GPUs in a cluster configuration. The graph below shows the solver time for each iteration for two unique ANSYS Fluent models. Both models utilize a GPU which results in faster completion of the solver.

Learn more about ANSYS by checking out:

• Accelerated ANSYS Fluent: Algebraic Multigrid on a GPU.
• ANSYS Improvements to Engineering Productivity with HPC and GPU-Accelerated Simulation

Check out a demonstration that provides a preview of future multi-GPU capability in this CPU-only vs. CPU+GPU simulation of an external aerodynamic solution for an automotive sedan. (Fluent demo embedded)