Re: [OMPI devel] Poor performance when compiling with --disable-dlopen

2018-01-23 Thread Gilles Gouaillardet
Dave, i can reproduce the issue with btl/openib and the IMB benchmark, that is known to MPI_Init_thread(MPI_THREAD_MULTIPLE) note performance is ok with OSU benchmark that does not require MPI_THREAD_MULTIPLE Cheers, Gilles On Wed, Jan 24, 2018 at 1:16 PM, Gilles Gouaillardet

Re: [OMPI devel] Poor performance when compiling with --disable-dlopen

2018-01-23 Thread Dave Turner
Gilles, I'm using NetPIPE which is available at http://netpipe.cs.ksu.edu My base test is uni-directional with 1 process on a node communicating with a process on a second node. make mpi mpirun -np 2 --hostfile=hf.2p2n NPmpi cat hf.2p2n node0 slots=1 node1 slots=1 NetPIPE does not do any

Re: [OMPI devel] Poor performance when compiling with --disable-dlopen

2018-01-23 Thread Gilles Gouaillardet
Dave, here is what I found - MPI_THREAD_MULTIPLE is not part of the equation (I just found it is no more required by IMB by default) - patcher/overwrite is not built when Open MPI is configure'd with --disable-dlopen - when configure'd without --disable-dlopen, performances are way worst for

Re: [OMPI devel] Poor performance when compiling with --disable-dlopen

2018-01-23 Thread Paul Hargrove
Ah, this sounds familiar. I believe that the issue Dave sees is that without patcher/overwrite the "leave pinned" protocol is OFF by default. Use of '-mca mpi_leave_pinned 1' may help if my guess is right. HOWEVER, w/o the memory management hooks provided using patcher/overwrite, leave pinned

Re: [OMPI devel] Poor performance when compiling with --disable-dlopen

2018-01-23 Thread Gilles Gouaillardet
Thanks Paul, unfortunately, that did not help :-( performance are just as bad even with --mca mpi_leave_pinned 1 and surprisingly, when patcher/overwrite is used, performances are not worse with --mca mpi_leave_pinned 0 Cheers, Gilles On Wed, Jan 24, 2018 at 2:28 PM, Paul Hargrove

[OMPI devel] Open MPI 3.0.1rc2 available for testing

2018-01-23 Thread Barrett, Brian via devel
I’ve posted the first public release candidate of Open MPI 3.0.1 this evening. It can be downloaded for testing from: https://www.open-mpi.org/software/ompi/v3.0/ We appreciate any testing you can do in preparation for a release in the next week or two. Thanks, Brian & Howard

Re: [OMPI devel] Poor performance when compiling with --disable-dlopen

2018-01-23 Thread Gilles Gouaillardet
Dave, At first glance, that looks pretty odd, and I'll have a look at it. Which benchmark are you using to measure the bandwidth ? Does your benchmark MPI_Init_thread(MPI_THREAD_MULTIPLE) ? Have you tried without --enable-mpi-thread-multiple ? Cheers, Gilles On Wed, Jan 24, 2018 at 12:55 PM,

Re: [OMPI devel] Poor performance when compiling with --disable-dlopen

2018-01-23 Thread Gilles Gouaillardet
Dave, one more question, are you running the openib/btl ? or other libraries such as MXM or UCX ? Cheers, Gilles On 1/24/2018 12:55 PM, Dave Turner wrote:    We compiled OpenMPI 2.1.1 using the EasyBuild configuration for CentOS as below and tested on Mellanox QDR hardware.

[OMPI devel] Open MPI 3.1.0 pre-release available

2018-01-23 Thread Barrett, Brian via devel
The Open MPI team is pleased to announce the first pre-release of the Open MPI 3.1 series, available at: https://www.open-mpi.org/software/ompi/v3.1/ RC1 has two known issues: - We did not complete work to support hwloc 2.x, even when hwloc is built as an external library. This may or

[OMPI devel] Poor performance when compiling with --disable-dlopen

2018-01-23 Thread Dave Turner
We compiled OpenMPI 2.1.1 using the EasyBuild configuration for CentOS as below and tested on Mellanox QDR hardware. ./configure --prefix=/homes/daveturner/libs/openmpi-2.1.1c --enable-shared --enable-mpi-thread-multiple --with-verbs