Tesla Compute Cluster RRS feed

  • Question

  • It appears that Amazon AWS runs its Volta GPUs in Tesla Compute Cluster mode which does not expose a WDDM.

    I have lots of HPC code written for DirectCompute but the code is unable to find the GPU.

    Is this the end of DirectCompute and should I abandon it or will Microsoft come up with a solution to run DirectCompute in such an environment?

    Tuesday, May 22, 2018 11:27 PM


All replies