Hi Gilles,
I confirmed the vader is used when I don't specify any BTL as you pointed
out!
Regards,
Tetsuya Mishima
[mishima@manage OMB-3.1.1-openmpi2.0.0]$ mpirun -np 2 --mca
btl_base_verbose 10 -bind-to core -report-bindings osu_bw
[manage.cluster:20006] MCW rank 0 bound to socket 0[core 0[hwt
Hi,
Thanks. I will try it and report later.
Tetsuya Mishima
2016/07/27 9:20:28、"devel"さんは「Re: [OMPI devel] sm BTL performace of
the openmpi-2.0.0」で書きました
> sm is deprecated in 2.0.0 and will likely be removed in favor of vader in
2.1.0.
>
> This issue is probably this known issue:
Also, btl/vader has a higher exclusivity than btl/sm, so if you do not
manually specify any btl, vader should be used.
you can run with
--mca btl_base_verbose 10
to confirm which btl is used
Cheers,
Gilles
On 7/27/2016 9:20 AM, Nathan Hjelm wrote:
sm is deprecated in 2.0.0 and will
sm is deprecated in 2.0.0 and will likely be removed in favor of vader in 2.1.0.
This issue is probably this known issue:
https://github.com/open-mpi/ompi-release/pull/1250
Please apply those commits and see if it fixes the issue for you.
-Nathan
> On Jul 26, 2016, at 6:17 PM,
Hi Gilles,
Thanks. I ran again with --mca pml ob1 but I've got the same results as
below:
[mishima@manage OMB-3.1.1-openmpi2.0.0]$ mpirun -np 2 -mca pml ob1 -bind-to
core -report-bindings osu_bw
[manage.cluster:18142] MCW rank 0 bound to socket 0[core 0[hwt 0]]:
[B/././././.][./././././.]
Hi,
can you please run again with
--mca pml ob1
if Open MPI was built with mxm support, pml/cm and mtl/mxm are used
instead of pml/ob1 and btl/openib
Cheers,
Gilles
On 7/27/2016 8:56 AM, tmish...@jcity.maeda.co.jp wrote:
Hi folks,
I saw a performance degradation of openmpi-2.0.0
Hi folks,
I saw a performance degradation of openmpi-2.0.0 when I ran our application
on a node (12cores). So I did 4 tests using osu_bw as below:
1: mpirun –np 2 osu_bw bad(30% of test2)
2: mpirun –np 2 –mca btl self,sm osu_bw good(same as openmpi1.10.3)
3:
Gilles,
With the additional information you provided about "dependency_libs", I
agree that that either of the fixes you propose sound safe.
-Paul
On Mon, Jul 25, 2016 at 6:26 PM, Gilles Gouaillardet
wrote:
> Paul,
>
> in my environment, libslurm.la contains
>
> # Linker
Hi Sreenidhi
Only partial resolution. By pushing out the eager path to 4 MB we were able to
get around 2GB/sec per socket connection with osu bw test.
The kernel is quite old though - 2.6.x - and being a summer student project
with a focus on IB vs rout able ROCE we've moved on.
Howard
Hi Gilles,
On 25/07/16 10:38 PM, Gilles Gouaillardet wrote:
Eric,
where can your test case be downloaded ? how many nodes and tasks do you
need to reproduce the bug ?
Sadly, it is in our in-house code and it requires to whole source code
which isn't public... :/
I have this bug with 20
hi Howard,
Was this issue resolved ? If so, what is the solution ?
Please let me know.
Curious to know , since we are also experimenting with these limits.
Thanks,
- Sreenidhi.
On Tue, Jul 19, 2016 at 10:50 AM, Gilles Gouaillardet
wrote:
> Howard,
>
>
> did you bump both
Nathan and Eric,
there is a know issue of libnbc not correctly retaining datatypes, for
example, if you start a non blocking collective operation (MPI_Ibcast
for example) and they MPI_Type_destroy() the datatype *before* the non
blocking collective completes, then the datatype is free'd by
12 matches
Mail list logo