On Wed, May 05, 2010 at 06:08:57PM -0400, Gus Correa wrote:
> If anybody else has Open MPI working with hyperthreading and "sm"
> on a Nehalem box, I would appreciate any information about the
> Linux distro and kernel version being used.
Debian 5 (lenny), Core i7 920, Asus P6T MoBo, 12GB RAM, Op
Hi Ralph
Thank you.
Yes, I will give the user/researcher that TCP solution for now,
because he needs to start running his model with Open MPI.
He bought a brand new super-duper machine, with two-way Nehalem,
48GB RAM, etc, and so far he couldn't do any work,
which is frustrating.
I googled aroun
I saw similar issues in my former life when we encountered a Linux "glitch" in
the way it handled proximity for shared memory - caused lockups under certain
conditions. Turned out the problem was fixed in a later kernel version.
Afraid I can't remember the versions involved any more, though
Ah, missed that - afraid I no speakee fortran any more (thankfully got to
remove that module from my brain 20+ years ago).
On May 5, 2010, at 1:18 PM, Andrew J Marquis wrote:
> Dear Ralph,
>
> thanks for that. I have done much the same (as I indicated in my original
> post). I this case my C-
Dear Ralph,
thanks for that. I have done much the same (as I indicated in my original
post). I this case my C-program correctly spawned the slaves and the slaves
printed the correctly passed argument lists. On running this and my fortran
slave I get:
nsize, mytid: iargs 2
I think OMPI is okay - here is a C sample program and the associated output:
$ mpirun -np 3 ./spawn_multiple
Parent [pid 98895] about to spawn!
Parent [pid 98896] about to spawn!
Parent [pid 98897] about to spawn!
Parent done with spawn
Parent sending message to children
Parent done with spawn
Par
Hi,
I am using mpi_comm_spawn_multiple to spawn multiple commands with argument
lists. I am trying to do this in fortran (77) using version openmpi-1.4.1 and
the ifort compiler v9.0. The operating system is SuSE Linux 10.1 (x86-64).
I have put together a simple controlling example program (te
Hi Jeff, Ralph, list.
Sorry for the long email, and the delay to answer.
I had to test MPI/reboot the machine several times
to address the questions.
Hopefully with answers to all your questions inline below.
Jeff Squyres wrote:
I'd actually be a little surprised if HT was the problem.
I run w
In general, even in your serial fortran code, you're already taking a
performance hit using a derived type. Is it really necessary? It might be
easier for you to change your fortran code into more memory friendly structures
and then the MPI part will be easier. The serial code will have the adde
On May 5, 2010, at 9:48 AM, Gus Correa wrote:
> Jeff: Should I wait for the 1.0 stable release?
> Use the current stable v0.9.3?
> John says one can go fearless with the 1.0 candidate release,
> but I tend to chicken out at cutting edge stuff.
1.0rc4 is *really* close to the final. Samuel just m
Douglas Guptill wrote:
> On Tue, May 04, 2010 at 05:34:40PM -0600, Ralph Castain wrote:
>> On May 4, 2010, at 4:51 PM, Gus Correa wrote:
>>
>>> Hi Ralph
>>>
>>> Ralph Castain wrote:
One possibility is that the sm btl might not like that you have
hyperthreading enabled.
>>> I remember tha
Gus Correa wrote:
> Hi Ralph
>
> Ralph Castain wrote:
>> One possibility is that the sm btl might not like that you have
>> hyperthreading enabled.
>
> I remember that hyperthreading was discussed months ago,
> in the previous incarnation of this problem/thread/discussion on
> "Nehalem vs. Open M
Vedran Coralic wrote:
> Hello,
>
> In my Fortran 90 code I use several custom defined derived types.
> Amongst them is a vector of arrays, i.e. v(:)%f(:,:,:). I am wondering
> what the proper way of sending this data structure from one processor to
> another is. Is the best way to just restructure
Hi John and Jeff
John: Thank you very much for pointing this out.
I would never knew about it without your tip.
I have yet to understand what it does
(I'm browsing the documentation now), and how
it can help understand, and perhaps monitor and allocate,
CPU/core resources.
Jeff: Should I wait f
Yes, you can use derived datatypes in MPI -- but be sure to read the language
chapter in the MPI-2.2 spec to be aware of a series of issues with Fortran.
We're actively working on "better" Fortran MPI bindings that won't have issues
with sending Fortran derived types (the current "medium" size
On May 5, 2010, at 1:14 AM, John Hearns wrote:
> Regarding hyperthreading, and finding our information about your CPUs
> in detail, there is the excellent hwloc project from OpenMPI
>
> http://www.open-mpi.org/projects/hwloc/
>
> I downloaded the 1.0 release candidate, and it compiled and ran fi
Hi!
I'm using OpenMPI 1.3 on 30 nodes connected with Gigabit Ethernet on Redhat
Linux x86_64.
Our MPI job sometimes hang and show follow error logs:
[btl_tcp_frag.c:216:mca_btl_tcp_frag_recv] mca_btl_tcp_frag_recv: readv
failed: Connection timed out (110)
I run a test like this: wri
Regarding hyperthreading, and finding our information about your CPUs
in detail, there is the excellent hwloc project from OpenMPI
http://www.open-mpi.org/projects/hwloc/
I downloaded the 1.0 release candidate, and it compiled and ran first
time on Nehalem systems. Gives a superb and helpful view
18 matches
Mail list logo
