gpu - HyperQ support in OpenCL -


i want run heterogeneous kernels execute on single gpu asynchronously. think possible in nvidia kepler k20(or device having compute capability 3.5+) launching each of kernels different stream , runtime system maps them different hardware queues based on resource availability. feature accessible in opencl? if so, equivalent of cuda 'stream' in opencl? nvidia drivers support such execution on k20 cards through opencl? amd gpu has similar feature(or there on development)? answer of these questions me lot.

in principle, can use opencl command queues achieve cke (concurrent kernel execution). can launch them different cpu threads. here few links might started: how know if kernels executing concurrently? http://devgurus.amd.com/thread/142485

i not sure how work nvidia kepler gpus having strange issues using opencl on k20 gpu.


Comments

Popular posts from this blog

c# - DetailsView in ASP.Net - How to add another column on the side/add a control in each row? -

javascript - firefox memory leak -

Trying to import CSV file to a SQL Server database using asp.net and c# - can't find what I'm missing -