When I run parallel processors on a gpu card (different instances of one program), the speed drops. Is it always the case? Should I handle multi-threading in my code or there is a way that I can run different processes in parallel?
Related topics
| Topic | Replies | Views | Activity | |
|---|---|---|---|---|
| Parallel computing question | 3 | 4509 | June 3, 2011 | |
| different type of multiple GPUs one one system | 2 | 3255 | April 17, 2008 | |
| How to use CUDA in different CPU threads? | 1 | 1073 | December 17, 2009 | |
| A question the parallelization | 1 | 1183 | July 28, 2008 | |
| 32-256+ different process running in parallel | 3 | 3542 | August 4, 2009 | |
| Asynchronous processing? | 0 | 1672 | February 23, 2011 | |
| the possibility of two CUDA program run in a GPU | 1 | 883 | November 29, 2011 | |
| choosing which to parallelize | 2 | 3886 | February 1, 2009 | |
| Is it possible to take advantage of multi-core CPU parallelism while using CUDA? Considering using C | 3 | 5587 | April 4, 2011 | |
| Performance drop when using the processes using different gpus on one machine | 1 | 880 | March 28, 2015 |