In an effort to provide faster performance during solution, Ansys Polyflow supports offloading key solver computations onto a graphical processing unit (GPU) on a graphics card to accelerate those computations. Only high-end graphics cards, the ones with the most amount of cores and memory, will accelerate the solver computations with the best efficiency (warnings will be printed at the beginning and end of the listing file if the GPU is not efficient enough). The speedup provided by the GPU is a function of the CPU (that is, older / slower CPUs will benefit more than newer / faster ones), as well as the problem type and the size and configuration of the mesh.

It is important to understand that a GPU does not replace the CPU core(s) on which a simulation typically runs. One or more CPU cores must be used to run the Ansys Polyflow program. The GPUs are used in support of the CPU to process certain calculations. The CPU continues to handle most operations and will automatically offload some of the time-intensive parallel operations performed by certain equation solvers. These parallel solver operations can usually be performed much faster on the highly parallel architecture of a GPU, thus accelerating these solvers and reducing the overall time to solution.

On machines containing multiple GPU accelerator devices, the program automatically selects the GPU accelerator device to be used for the simulation. The program cannot detect if a GPU device is currently being used by other software, including another ANSYS simulation. Therefore, in a multi-user environment, you should be careful not to oversubscribe the GPU accelerator devices by simultaneously launching multiple simulations that attempt to use the same GPU to accelerate the solution. For more information, see Troubleshooting.

When running a simulation in parallel, it is possible that more than one of the CPUs could benefit from GPU support, though the GPU can only support one CPU at a time. By default, CPUs will proceed with their calculations unsupported if the GPU is already engaged. If you would prefer that CPUs wait for the GPU to become free, launch Polyflow from the command line with the -WaitGPU command line option.

For licensing, a GPU is considered as an additional core. Running on a CPU core + a GPU is equivalent to running on 2 cores. Running on 4 CPU cores + a GPU is equivalent to running on 5 cores.