1.7.1. Activating the GPU Accelerator Capability

The general procedure for using the GPU accelerator capability is as follows:

Before activating the GPU accelerator capability, you must have at least one supported GPU card with the proper driver level and proper HPC licensing.
Launch Polyflow with the GPU accelerator capability:
- If launching Polyflow from the command line, use the -acc command line option:
```
polyflow -acc nvidia -na N
```
  The -na command line option followed by a number (N) indicates the number of GPU accelerator devices to use for the simulation. If only the -acc option is specified, the program uses a single GPU device for the simulation by default (that is, -na 1). You may specify a value for N that is greater than 1.
  For parallel simulations (using the -th M command line option), you can add the -WaitGPU command line option if you want the CPUs to halt their calculations and wait for the GPU to be free if it is already being used by another CPU; otherwise, if the GPU is being used by a CPU, other CPUs will proceed unsupported.
  When launching a Polyflow calculation from Workbench, right-click the Solution cell, select Preferences..., and select Polyflow. In the Polyflow options... dialog box that opens, select nvidia from the GPU drop-down list.

With the GPU accelerator capability, the acceleration obtained by using the parallelism on the GPU hardware occurs only during the solution operations. Operational randomness and numerical round-off inherent to any parallel algorithm can cause slightly different results between runs on the same machine when using or not using the GPU hardware to accelerate the simulation.