[Paraview] Coprocessing with accelerators

[Vanmoer, Mark W]

Hello, for a CUDA application with coprocessing, is it correct that between a simulation step and coprocessing that the data will make a roundtrip from the device to the host and back to the device, since parts of the vis pipeline run on the CPU?

Can Piston work with coprocessing, would that avoid this round-trip?

How about for Xeon Phis?

Thanks,
Mark
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.paraview.org/pipermail/paraview/attachments/20130820/7b919dec/attachment-0001.htm>