Re: [OMPI users] unable to open osc pt2pt

2007-12-19 Thread de Almeida, Valmor F.
I don't know how to access the svn repository. svn checkout http://mpi4py.scipy.org/mpi4py/browser/mpi4py/trunk mpi4py svn: PROPFIND request failed on '/mpi4py/browser/mpi4py/trunk' svn: PROPFIND of '/mpi4py/browser/mpi4py/trunk': 200 Ok (http://mpi4py.scipy.org) Also signed up for the mpi4py us

Re: [OMPI users] Torque and OpenMPI 1.2

2007-12-19 Thread Ralph H Castain
It is fully implemented, but on my development branch at the moment. We hope to bring that over to the trunk late Jan - primarily need to complete some work on MPI-2 dynamic process management and give Josh a chance to repair the checkpoint/restart functionality before we bring it over. Ralph O

Re: [OMPI users] Torque and OpenMPI 1.2

2007-12-19 Thread Adams, Brian M
Ralph, Thanks for the clarification as I'm dealing with workarounds for this at Sandia as well... I might have missed this earlier in the dialog, but is this capability in the SVN trunk right now, or still on the TODO list? Brian Brian M. Adams, PhD (bria

Re: [OMPI users] Torque and OpenMPI 1.2

2007-12-19 Thread Ralph H Castain
Open MPI 1.3 will support use of the hostfile and the tm launcher simultaneously. It will work slightly differently, though, with respect to the hostfile: 1. PBS_NODEFILE will be read to obtain a complete list of what has been allocated to us 2. you will be allowed to provide a hostfile for each

Re: [OMPI users] Torque and OpenMPI 1.2

2007-12-19 Thread pat . o'bryant
Ralph, Thanks for the information. I am assuming OpenMPI 1.3 will support the "-hostfile" without the extra parms. Will 1.3 also carry the same restrictions you list below? Pat J.W. (Pat) O'Bryant,Jr. Business Line Infrastructure Technical Systems, HPC Office: 713-431-7022

[OMPI users] Fwd: [all-osl-users] Outage of the OSL services

2007-12-19 Thread Jeff Squyres
FYI -- all of Open MPI's electronic resources will be offline for half of Dec 26 for some scheduled maintenance. Please see below for details. Begin forwarded message: From: DongInn Kim Date: December 19, 2007 12:13:48 PM EST To: all-osl-us...@osl.iu.edu Subject: [all-osl-users] Outage of

Re: [OMPI users] new to MPI+infiniband

2007-12-19 Thread Brian Budge
Wow, I'm amazed... it just works "out-of-the-box" configured with with-openib. I guess the new kernel supplies everything I need. Thanks a bunch for the clues. Brian On Dec 19, 2007 5:55 AM, Jeff Squyres wrote: > In general, you need a functioning IB subsystem on your machines for > OMPI to

Re: [OMPI users] error with Vprotocol pessimist

2007-12-19 Thread Thomas Ropars
Hi, I have the same error message when fault tolerance is activated. I'm using gcc version 4.1.3, with Ubuntu 7.10 (i686) (kernel 2.6.22-14-generic) Thomas Aurelien Bouteiller wrote: If you want to use the pessimist message logging you have to use the "- mca vprotocol pessimist" flag on your

Re: [OMPI users] Theoretical Differences: Cluster Computing, Distributed Computing, Grid Computing

2007-12-19 Thread Amit Kumar Saha
On 12/19/07, Jeff Squyres wrote: > On Dec 19, 2007, at 11:33 AM, Amit Kumar Saha wrote: > > > This probably is the same as saying that : In Distributed computing, > > the member nodes are *not* dedicated machines solely for executing the > > specific HPC task, where as in a Cluster we have a dedic

Re: [OMPI users] Theoretical Differences: Cluster Computing, Distributed Computing, Grid Computing

2007-12-19 Thread Jeff Squyres
On Dec 19, 2007, at 11:33 AM, Amit Kumar Saha wrote: This probably is the same as saying that : In Distributed computing, the member nodes are *not* dedicated machines solely for executing the specific HPC task, where as in a Cluster we have a dedicated set of nodes, specifically working on a pa

Re: [OMPI users] Theoretical Differences: Cluster Computing, Distributed Computing, Grid Computing

2007-12-19 Thread Amit Kumar Saha
On 12/19/07, Aurelien Bouteiller wrote: > I have quite different definitions than Jeff. > > Distributed computing is encompassing all the "parallel computing" > models, including clusters, grids, master-slave, shared memory... > Everything that basically implies using several collaborating > proce

Re: [OMPI users] Theoretical Differences: Cluster Computing, Distributed Computing, Grid Computing

2007-12-19 Thread Amit Kumar Saha
On 12/19/07, Jeff Squyres wrote: > It's a pretty hazy difference; I don't think there are formal > definitions for each. > > Cluster computing *usually* (but not always) implies a bit more > tightly coupled set of computing: the app communicates and coordinates > between itself more than your stan

Re: [OMPI users] Torque and OpenMPI 1.2

2007-12-19 Thread Ralph H Castain
Just to be clear: what this does is tell Open MPI to launch using the SSH launcher. This will work okay, but means that Torque doesn't know about the children and cannot monitor them. It also won't work on clusters (such as the ones we have here) that do not allow you to ssh procs onto the backend

Re: [OMPI users] Theoretical Differences: Cluster Computing, Distributed Computing, Grid Computing

2007-12-19 Thread Aurelien Bouteiller
I have quite different definitions than Jeff. Distributed computing is encompassing all the "parallel computing" models, including clusters, grids, master-slave, shared memory... Everything that basically implies using several collaborating processes to solve a problem (whatever collaborat

Re: [OMPI users] Torque and OpenMPI 1.2

2007-12-19 Thread pat . o'bryant
Andy, I second the praise of the developers. The developers along with an involved user community can lead to nothing but success. Pat J.W. (Pat) O'Bryant,Jr. Business Line Infrastructure Technical Systems, HPC Office: 713-431-7022 "Caird, Andrew

Re: [OMPI users] Torque and OpenMPI 1.2

2007-12-19 Thread Caird, Andrew J
Glad to hear that worked for you. Full credit goes to Brock Palen who told me about this. It turns out we also have a user who wanted to do that. And meta-credit goes to the OMPI developers for making a consistent and flexible set of MPI tools and libraries. --andy > -Original Message-

Re: [OMPI users] Torque and OpenMPI 1.2

2007-12-19 Thread pat . o'bryant
Andrew, That worked like a champ. Now my users can have it both ways. For the record, my control statements looked like the following: /opt/openmpi-1.2.4/bin/mpirun -mca pls ^tm -np $NP -hostfile $PBS_NODEFILE $my_binary_path My job works just fine and reports no errors. This version of Open

Re: [OMPI users] Theoretical Differences: Cluster Computing, Distributed Computing, Grid Computing

2007-12-19 Thread Jeff Squyres
It's a pretty hazy difference; I don't think there are formal definitions for each. Cluster computing *usually* (but not always) implies a bit more tightly coupled set of computing: the app communicates and coordinates between itself more than your standard manager/worker computation mode

Re: [OMPI users] Torque and OpenMPI 1.2

2007-12-19 Thread Caird, Andrew J
oops, I meant -mca, not -mcs > -Original Message- > From: users-boun...@open-mpi.org > [mailto:users-boun...@open-mpi.org] On Behalf Of Caird, Andrew J > Sent: Wednesday, December 19, 2007 8:57 AM > To: Open MPI Users > Cc: users-boun...@open-mpi.org > Subject: Re: [OMPI users] Torque an

Re: [OMPI users] Torque and OpenMPI 1.2

2007-12-19 Thread Caird, Andrew J
Does OMPI built with TM but run with: -mcs pls ^tm give the same effect? --andy > -Original Message- > From: users-boun...@open-mpi.org > [mailto:users-boun...@open-mpi.org] On Behalf Of > pat.o'bry...@exxonmobil.com > Sent: Wednesday, December 19, 2007 8:47 AM > To: Open MPI Users

Re: [OMPI users] new to MPI+infiniband

2007-12-19 Thread Jeff Squyres
In general, you need a functioning IB subsystem on your machines for OMPI to build for and use IB. This includes both the set of kernel drivers and the userspace libraries. In particular, Open MPI v1.2 uses the "ibverbs" library. If OMPI can find the ibverbs library and headers during it

Re: [OMPI users] new to MPI+infiniband

2007-12-19 Thread pat . o'bryant
Brian, Here is how I do it: ./configure --prefix /opt/openmpi-1.2.4 --with-openib=/usr/local/ofed \ --without-tm CC=icc CXX=icpc F77=ifort FC=ifort \ --with-threads=posix --enable-mpi-threads J.W. (Pat) O'Bryant,Jr. Business Line Infrastructure Technical Systems, HPC Office: 713-431-7022

Re: [OMPI users] Torque and OpenMPI 1.2

2007-12-19 Thread pat . o'bryant
Terry, Your suggestion worked. So long as I specifically state "--without-tm", the OpenMPI 1.2.4 build allows the use of "-hostfile". Apparently, by default, OpenMPI 1.2.4 will incorporate Torque if it exists, so it is necessary to specifically request "no Torque support". I used the normal To

[OMPI users] new to MPI+infiniband

2007-12-19 Thread Brian Budge
Hi all - I have been using OpenMPI for quite a while now, and its working out great. I was looking at the FAQ and trying to figure out how to configure OpenMPI with infiniband. It shows how to enable IB pointing to the OFED directory. I have infiniband built into the kernel, along with IP over I

[OMPI users] Theoretical Differences: Cluster Computing, Distributed Computing, Grid Computing

2007-12-19 Thread Amit Kumar Saha
Hi all! I am a bit hazy about the differences between Distributed Computing and Cluster Computing. This document here http://www.ibm.com/developerworks/grid/library/gr-heritage/ is a nice read on Grid Computing and it also has few lines about important differences from Cluster Computing. Hints w