CST STUDIO SUITE GPU ComputingSource: CST of America, Inc.
The high memory bandwidth and parallel processing abilities of GPU cards mean that GPU computing can provide signifi cant simulation speed advantages over conventional CPU computing. CST STUDIO SUITE® supports GPU computing in the transient solver, integral equation solver and particle-in-cell (PIC) solver, using NVIDIA Tesla® GPU cards.
The GPU Computing Guide, available on our website, gives a complete, up-to-date list of supported hardware. With the transient solver and integral equation solver in CST MICROWAVE STUDIO® (CST MWS®), up to eight GPUs can be used per host system, with additional GPUs conferring further performance improvements. All GPU-capable solvers support the newest generation of Tesla cards, Kepler, which offers a performance improvement of around 30% over the previous generation, Fermi.