Just to clarify: are you talking about using this on RoadRunner to
move data from the Opteron to the Cell blades? Or are you talking
about using OpenMPI on some other machine?
If you are interested in RR, we should probably continue the
discussion off-list. The short answer is that we have a mechanism for
launching MPI procs on the cell blades, and there has indeed been work
done to create a BTL so that OMPI can move data across between the
Opterons and cells. However, it has never been added to the OMPI code
base and/or fully completed to production quality.
If there is interest in such capability, it can be done. We just
hadn't heard any expressions of interest, and so it remains on the
On Dec 11, 2008, at 5:01 PM, Marcus G. Daniels wrote:
> I've heard some experimental work has been done to run OpenMPI over
> the Axon driver as found in IBM triblades. Seems like that should
> work fine, as it's just another RDMA interface, no? I'd like to use
> MPI for features like derived types, and moving around non-flat data.
> Marcus Daniels
> users mailing list