Enable GPU acceleration for Sibernetic running within stack
slarson opened this issue · 2 comments
slarson commented
Currently if OpenCL doesn't see the local GPU, it just falls back to the CPU, which makes the stack easily usable by everyone but makes it more challenging to take advantage of high performance. However, we would like to get it working to recognize a local GPU.
We have attempted to do this but drivers on the host need to be set up correctly to enable this and so it adds an additional dependency on the quickstart process.
We'd like to have an easy way to set this up, whether it is an intel GPU, an NVIDIA GPU, or what have you.
clausagerskov commented
running within stack means within docker?
mwatts15 commented
@clausagerskov yes.