I hope I'm still on the right list for my current problem.
Today we figured out on a similiar but older four opteron (6100) 48
cores system that
mpiexec -bind-to numa is the essential key point.
This I want to realize on my system. I already installed libnuma such
that hwloc configure
uses numa.
Then I configured openmpi-1.10.0 which also uses libnuma
When I compile my petsc example with MPIEXEC="orterun -bind-to numa"
and run the application
I get
---------------------
A request was made to bind a process, but at least one node does NOT
support binding processes to cpus.
Node: leo
This usually is due to not having libnumactl and libnumactl-devel
installed on the node.
---------------------
I cannot find these packages for ubuntu 14.04.
Even when I compile numactl-2.0.9 from
http://oss.sgi.com/projects/libnuma/
I only generates libnuma
I find a lot of ubuntu deb packages on
https://launchpad.net/ubuntu/+source/numactl
But there I only find libnuma but no libnumactl.
Where do I get the libnumactl and libnumactl-devel from?
Is this the wrong thread and the wrong list?
I have a feeling that I'm quite close but just cannot reach it :(
Thanks,
Fabian
On 10/27/2015 04:05 PM, Brice Goglin wrote:
I guess the next step would be to look at how these tasks are placed on
the machine. There are 8 NUMA nodes on the machine. Maybe 9 is where it
starts placing a second task per NUMA node?
For OMPI, --report-bindings may help. I am not sure about MPICH.
Brice
Le 27/10/2015 15:52, Fabian Wein a écrit :
On 10/27/2015 03:42 PM, Brice Goglin wrote:
I guess the problem is that your OMPI uses an old hwloc internally. That
one may be too old to understand recent XML exports.
Try replacing "Package" with "Socket" everywhere in the XML file.
Thanks! That was it.
I now get almost perfectly reproducible results.
np speedup
1 1.0
2 1.99
3 2.98
4 3.98
5 4.89
6 5.9
7 6.89
8 7.87
9 5.44
10 6.04
11 6.55
12 7.0
13 7.75
14 8.24
15 8.41
16 9.4
17 7.33
18 7.16
19 8.05
20 8.39
What still puzzles me is the almost perfect speedup up to eight and
than the
drop down. But for the beginning 8 is already good!
Thanks again,
Fabian
_______________________________________________
hwloc-users mailing list
hwloc-us...@open-mpi.org
Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/hwloc-users
Link to this post:
http://www.open-mpi.org/community/lists/hwloc-users/2015/10/1210.php
_______________________________________________
hwloc-users mailing list
hwloc-us...@open-mpi.org
Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/hwloc-users
Link to this post:
http://www.open-mpi.org/community/lists/hwloc-users/2015/10/1210.php
