I know the Tesla servers are intended mainly for Cuda, but could they also work for DirectX HLSL programs?
We have some code running now in HLSL under directx that will take some time to port to Cuda, but would like to benefit from the rackmount form factor sooner.
Regards,
LTG