As far as I can see the application runs using OpenGL and from what I have seen the Tesla GPUs are compatible with it however it would be good to know if anyone has tried using Tesla in a thin client environment before.
Tesla cards aren’t GPUs in the conventional sense, in that they can’t drive a display. They are solely intended for numerical computation. To the best of my knowledge, SolidWorks doesn’t support CUDA at all, so a Tesla card would be completely useless in such an application.
If you are looking for some sort of deskside GPU for OpenGL applications that need a lot of memory, something like a Quadro Plex D2 might be closer to what you probably want. I haven’t seen many “thin” clients with PCI-e x16 connectors that are going to be suitable for driver high performance video cards, though. What client dod you have (or are planning to buy)?
I was thinking more of using the Tesla attached to the Terminal Server so that the Tesla would perform the graphics processing of the server (compared to the Thin client performing the graphics processing).
Although SolidWorks isnt CUDA compatible it is based on OpenGL which I belive the Tesla supports.