On 7/9/07, Jeff Squyres wrote:
On Jul 6, 2007, at 12:05 PM, Alex Tumanov wrote:
> Eureka! I managed to get it working despite the incorrect _initial_
> ./configure invocation. For those interested, here are my compilation
> options:
> # cat ompi_build.sh
> #!/bin/sh
>
> r
uot;_defaultdocdir %{_prefix}/share" \
-D "mflags all" openmpi-1.2.3-1.src.rpm
Thanks for looking :)
Alex.
On 7/5/07, Alex Tumanov wrote:
Actually, tried compiling the RPM again, and at the very top, noticed
that the ./configure is called with --sysconfdir set to /opt/o
Actually, tried compiling the RPM again, and at the very top, noticed
that the ./configure is called with --sysconfdir set to /opt/openmpi
instead of the new name provided. All other parameters are correct!
Any ideas?
./configure --build=x86_64-redhat-linux-gnu
--host=x86_64-redhat-linux-gnu --ta
Greetings,
The spec file provided in the latest stable src RPM makes it possible
to change the name of the resulting RPM. I tried to make use of that,
but ran into some issues. Specifically, the resulting RPM does not
have the etc directory (and sample config files in it). rpmbuild
complained abo
forgive a trivial question, but what's a multi-rail?
On 6/8/07, George Bosilca wrote:
A fix for this problem is now available on the trunk. Please use any
revision after 14963 and your problem will vanish [I hope!]. There
are now some additional parameters which allow you to select which
Myrine
Greetings,
Very quick question: I noticed that liborte and libopal shared
libraries are no longer present in the 1.2 version of Open MPI. Can I
establish backward compatibility symlinks for applications that were
compiled with Open MPI 1.1.4 and were linked with liborte and libopal?
If so , which
nies, which probably do not have
any vested interest in making their interconnects synergistically
coexist with the ones of their competitors or with interconnects the
companies are dropping support for.
Many thanks,
Alex.
On Feb 12, 2007, at 6:48 PM, Alex Tumanov wrote:
> Anyone else
Hello,
I recently tried running HPLinpack, compiled with OMPI, over myrinet
MX interconnect. Running a simple hello world program works, but XHPL
fails with an error occurring when it tries to MPI_Send:
# mpirun -np 4 -H l0-0,c0-2 --prefix $MPIHOME --mca btl mx,self
/opt/hpl/openmpi-hpl/bin/xhpl
.
On 2/8/07, Alex Tumanov wrote:
Thanks for your insight George.
> Strange, the latency is supposed to be there too. Anyway, the latency
> is only used to determine which one is faster, in order to use it for
> small messages.
I searched the code base for mca parameter registering
mpirun --prefix /opt/openmpi -mca oob_tcp_include eth0 -mca
btl_tcp_if_include eth0 --hostfile ~/work/openmpi_hostfile -np 4
hostname
Could a section be added to the FAQ mentioning that the firewall
service should be shutdown over the mpi interface and that the two
-mca switches should be used?
I have added the following line to my .bashrc:
export OMPIFLAGS="-mca oob_tcp_include eth0 -mca btl_tcp_if_include
eth0 --hostfile ~/work/openmpi_hostfile"
and have verified that mpirun $OMPIFLAGS -np 4 hostname works.
Is there a better way of accomplishing this, or is this a matter of
there be
Hello Troy,
This issue is fairly common and has to do with the maximum amount of
memory allowed to be allocated. See this FAQ for more detail:
http://www.open-mpi.org/faq/?category=openfabrics#ib-locked-pages
I had the same issue and the FAQ resolves it.
Good luck,
Alex.
On 2/8/07, Troy Telfor
gt; btl_tcp_if_include = eth0,eth1
>> btl_tcp_max_rdma_size = 524288
>>
>> btl_tcp_latency_eth0 = 47
>> btl_tcp_bandwidth_eth0 = 587
>>
>> btl_tcp_latency_eth1 = 51
>> btl_tcp_bandwidth_eth1 = 233
>>
>> Something similar has to be done for op
p_latency_eth0 = 47
btl_tcp_bandwidth_eth0 = 587
btl_tcp_latency_eth1 = 51
btl_tcp_bandwidth_eth1 = 233
Something similar has to be done for openib and gm, in order to allow
us to strip the messages correctly.
Thanks,
george.
On Feb 8, 2007, at 12:02 PM, Alex Tumanov wrote:
> Hel
Hello Jeff. Thanks for pointing out NetPipe to me. I've played around
with it a little in hope to see clear evidence/effect of message
striping in OpenMPI. Unfortunately, what I saw is that the result of
running NPmpi over several interconnects is identical to running it
over a single fastest one
Hello,
mpirun -np 2 myprogram inputfile >outputfile
There can be a whole host of issues with the way you run your
executable and/or the way you have the environment setup. First of
all, when you ssh into the node, does the environment automatically
get updated with correct Open MPI paths? I.e.
Thanks for your reply, Jeff.
> It never occurred to me that the headnode would try to communicate
> with the slave using infiniband interfaces... Orthogonally, what are
The problem here is that since your IB IP addresses are
"public" (meaning that they're not in the IETF defined ranges for
priv
Thank you for your reply, Reese!
What version of GM are you running?
# rpm -qa |egrep "^gm-[0-9]+|^gm-devel"
gm-2.0.24-1
gm-devel-2.0.24-1
Is this too old?
And are you sure that gm_board_info
shows all the nodes that are listed in your machine file?
Yes, that was the issue - bad cable connec
Hello list,
I'm having difficulty with running a simple hello world OpenMPI
program over Myrinet gm interconnect - please see the log at the end
of this email. The error is tripped by a call to the function
gm_global_id_to_node_id(
gm_btl->port,
gm_endpoint->endpoint_addr.global_id,
That really did fix it, George:
# mpirun --prefix $MPIHOME -hostfile ~/testdir/hosts --mca btl
tcp,self --mca btl_tcp_if_exclude ib0,ib1 ~/testdir/hello
Hello from Alex' MPI test program
Process 0 on dr11.lsf.platform.com out of 2
Hello from Alex' MPI test program
Process 1 on compute-0-0.local o
nodes?
- Galen
On Feb 1, 2007, at 2:50 PM, Alex Tumanov wrote:
> Hi,
>
> I have kept doing my own investigation and recompiled OpenMPI to have
> only the barebones functionality with no support for any interconnects
> other than ethernet:
> # rpmbuild --rebuild --define="
replicate the error, I ran
# mpirun -hostfile ~/testdir/hosts --mca btl tcp,self ~/testdir/hello
In this case, you can even omit the runtime mca param specifications:
# mpirun -hostfile ~/testdir/hosts ~/testdir/hello
Thanks for reading this. I hope I've provided enough information.
Sincerel
Hello,
I have tried a very basic test on a 2 node "cluster" consisting of 2
dell boxes. One of them is dual CPU Intel(R) Xeon(TM) CPU 2.80GHz with
1GB of RAM and the slave node is quad-CPU Intel(R) Xeon(TM) CPU
3.40GHz with 2GB of RAM. Both have Infiniband cards and Gig-E. The
slave node is conne
23 matches
Mail list logo
