Hi,
I have 4 intel xeon quad core quad socket nodes connected via infiniband. This setup is currently used to rum MPI applications. Now I want to add an ATI card on each node and run a test OpenCL program. The program shall spawn across all nodes. For this puropose which MPI shall I use, or do I need MPI at all?
PS: It is necessary for me to run my OpenCL program over a cluster, and I am totally new to OpenCL.
I dont know if the question I asked is too trivial or I could not make my point clear. I currently have no hands on with OpenCL, but if it can be run over a cluster then I will have to learn it.
So, for now its a feasibilty study for me. Please help me.
Thanks.
OpenCL is API so whcih MPI implementation you useis up to you.
There is no reason why OpenCL and MPI would not work well together. I haven't actually tried it, but it certainly should work without any problems.
MPICH2 is a pretty popular and free MPI implementation for both Windows and Linux. I haven't had any problems with it on Windows at least.
Thanks nou and dravisher.
So, now I will try to get a couple of cards and write a helloworld application.