I do not know if here is the right place for this question.
I am using Python and PyCuda. If I create two threads in Python and both those threads call Kernels in Cuda through PyCuda how those kernels will be running?
If at same time, concurrently or serialized? The resources of the GPU are shared or not. If so the resources are not enough for both kernels. Must I do code to serialize the kernels? Which is the best way to do it?