Re: [OMPI users] multipath support for infiniband

2010-09-22 Thread Jeff Squyres
On Sep 22, 2010, at 8:04 AM, Jens Domke wrote: > I already tried this parameter, but I don't see any improvements in the > benchmarks. Additionally while doing further investigations into the opensm I > didn't see the QP requests for other LIDs than the base LIDs. I'm afraid that I'm not an Inf

Re: [OMPI users] multipath support for infiniband

2010-09-22 Thread Jens Domke
I already tried this parameter, but I don't see any improvements in the benchmarks. Additionally while doing further investigations into the opensm I didn't see the QP requests for other LIDs than the base LIDs. Regards Jens Jeff Squyres wrote: Yes, check out the btl_openib_max_lmc MCA param

Re: [OMPI users] multipath support for infiniband

2010-09-22 Thread Jeff Squyres
Yes, check out the btl_openib_max_lmc MCA parameter: shell$ ompi_info --param btl openib --parsable | grep lmc On Sep 21, 2010, at 11:45 AM, Jens Domke wrote: > Hello, > > the InfiniBand architecture has a LMC feature to assign mutiple virtual LIDs > to one port and so provides multiple paths

[OMPI users] multipath support for infiniband

2010-09-21 Thread Jens Domke
Hello, the InfiniBand architecture has a LMC feature to assign mutiple virtual LIDs to one port and so provides multiple paths between two ports. Is there a methode in openmpi to enable message-striping over these paths to increase bandwidth or avoid congestions? (I don't mean the multirail fe