Thanks to everyone who attended today's Open MPI State of the Union BOF session!
We started the recording a few minutes late (sorry about that!), but if you're
watching the video, know that you only missed an intro slide or two. I
*believe* that the SC people told us that the Zoom video
On Nov 15, 2021, at 2:20 PM, Jeff Squyres (jsquyres) via users
mailto:users@lists.open-mpi.org>> wrote:
SC'21 is hybrid this year; all the academic sessions and BOF are available to
online attendees.
We hope you'll join us tomorrow, 16 Nov 2021, at 1:15pm US Eastern time ... for
the
SC'21 is hybrid this year; all the academic sessions and BOF are available to
online attendees.
We hope you'll join us tomorrow, 16 Nov 2021, at 1:15pm US Eastern time /
2:15pm US Central time (i.e., local SC'21 time) for the annual Open MPI State
of the Union Birds of a Feather (BOF) session.
Gilles' question is correct; the larger point to make here: the openib BTL is
obsolete and is actively replaced by the UCX PML. UCX is supported by the
vendor (NVIDIA); openib is not.
If you're just starting a new project, I would strongly advocate using UCX
instead of openib.
On Nov 1,
Ray --
Looks like this is a dup of https://github.com/open-mpi/ompi/issues/8919.
On Sep 29, 2021, at 10:47 AM, Ray Muno via users
mailto:users@lists.open-mpi.org>> wrote:
Looking to compile
OpenMPI 4.1.1
under Centos 7.9, (with Mellanox OFED 5.3 stack)
with the nVidia HPC-SDK, version
This appears to be a legit bug with the use of MPI_T in the test/example
monitoring app, so I'm going to move the discussion to the Github issue so that
we can track it properly:
https://github.com/open-mpi/ompi/issues/9260
To answer Jong's question: ob1 is one of Open MPI's point-to-point
Yeah, it looks like orte-clean is busted in 4.0.x and 4.1.x. I have filed
https://github.com/open-mpi/ompi/issues/9171 to track the issue.
FWIW, orte-clean shouldn't be necessary for normal Open MPI runs. We should
fix it (or remove it), of course, but this shouldn't be a blocker for whatever
On Jun 18, 2021, at 10:30 AM, Mehdi Jenab via users
mailto:users@lists.open-mpi.org>> wrote:
I wonder if there is any way to send and receive a bitset in OpenMPI without
converting it to integer.
It depends on what type your bitset is. If it's a constant memory layout
(e.g., a fixed
John --
Can you supply all the information requested here:
https://www.open-mpi.org/community/help/
Thanks.
On May 25, 2021, at 5:44 PM, John Haiducek via users
mailto:users@lists.open-mpi.org>> wrote:
Hi,
When attempting to build OpenMPI in a Fedora 34 Docker image I get the
Sharon --
I replied to your comment on the corresponding GitHub issue:
https://github.com/open-mpi/ompi/issues/5818#issuecomment-842531001.
On May 14, 2021, at 1:12 PM, Sharon Brunett via users
mailto:users@lists.open-mpi.org>> wrote:
Does Open MPI v 4.1.0 or v 4.1.1 support using IP
FYI: general Open MPI questions are better sent to the user's mailing list.
Up through the v4.1.x series, the "orted" is a general helper process that Open
MPI uses on the back-end. It will not quit until all of its children have
died. Open MPI's run time is designed with the intent that some
You can rm $prefix/lib/openmpi/mca_btl_openib*
Or set the MCA param "btl" to "^openib" (which means: not openib) -- either in
the environment, on the command line, or in the global config file:
https://www.open-mpi.org/faq/?category=tuning#setting-mca-params
> On Apr 20, 2021, at 2:33 PM,
It's not really a problem, per se -- it's that Open MPI needs a lib event with
thread support enabled. The lib event that configure found / tried to use did
not have thread support enabled.
What was the command line you used to invoke configure?
> On Apr 11, 2021, at 1:52 PM, John Hearns via
rnelis
Networks.
On Wed, 7 Apr 2021 at 14:59, Jeff Squyres (jsquyres)
mailto:jsquy...@cisco.com>> wrote:
Check the output from ldd in a non-interactive login: your LD_LIBRARY_PATH
probably doesn't include the location of the Intel runtime.
E.g.
ssh othernode ldd /path/to/orted
Your s
Check the output from ldd in a non-interactive login: your LD_LIBRARY_PATH
probably doesn't include the location of the Intel runtime.
E.g.
ssh othernode ldd /path/to/orted
Your shell startup files may well differentiate between interactive and
non-interactive logins (i.e., it may set
Thanks to all who attended today. The slides from the presentation are now
available here:
https://www-lb.open-mpi.org/papers/ecp-bof-2021/
> On Mar 29, 2021, at 2:52 PM, Jeff Squyres (jsquyres) via announce
> wrote:
>
> Gentle reminder that the Open MPI State of the Union
of time!
Hope to see you there!
> On Mar 15, 2021, at 1:06 PM, Jeff Squyres (jsquyres)
> wrote:
>
> In conjunction with the Exascale Computing Project (ECP), George Bosilca,
> Jeff Squyres, and members of the Open MPI community will present the current
> status and future ro
In conjunction with the Exascale Computing Project (ECP), George Bosilca, Jeff
Squyres, and members of the Open MPI community will present the current status
and future roadmap for the Open MPI project.
We typically have an Open MPI "State of the Union" BOF at the annual
Supercomputing
On Mar 12, 2021, at 1:20 AM, Raut, S Biplab via users
mailto:users@lists.open-mpi.org>> wrote:
Reposting here without the logs – it seems there is a message size limit of
150KB and so could not attach the logs.
(Request the moderator to approve the original mail that has attachment of
What is the exact configure line you used to build Open MPI? You don't want to
put CC and CXX in a single quoted token. For example, do this:
./configure CC=gcc CXX=g++ ...
Don't do this (which is what your previous mail implied you might be doing...?):
./configure "CC=gcc CXX=g++" ...
On
want to know why OpenMPI
is not functioning as expected.
On Thu, 25 Feb 2021, 20:17 Jeff Squyres (jsquyres),
mailto:jsquy...@cisco.com>> wrote:
I don't know how many people on this list will be familiar with Termux or Arch
Linux. From a quick Google, it looks like Termux is an Android e
I don't know how many people on this list will be familiar with Termux or Arch
Linux. From a quick Google, it looks like Termux is an Android emulator (that
runs on Android? That doesn't make sense to me, but I'm wholly unfamiliar with
that ecosystem, so I don't have the background / grok the
4.1.1 is due "very soon" -- there's two outstanding blocker UCX issues that are
awaiting resolution.
That being said, you can certainly start with v4.1.0 today and upgrade to
v4.1.1 when it comes out. 4.1.1 will be ABI and backwards compatible with
v4.1.0..
On Feb 21, 2021, at 11:40 AM,
On Feb 17, 2021, at 12:24 PM, Luis Diego Pinto via users
wrote:
>
> Dear Christoph Niethammer,
>
> thank you very much for the advice, now i have imported the module "mpi_f08"
> and indeed those issues are fixed.
Exxxcellent. mpi_f08 is definitely the way to go for all new Fortran MPI
--Original Message-
>> From: users On Behalf Of Jeff Squyres
>> (jsquyres) via users
>> Sent: Tuesday, February 16, 2021 10:55 AM
>> To: Open MPI User's List
>> Cc: Jeff Squyres (jsquyres)
>> Subject: [EXTERNAL] Re: [OMPI users] mpi_f08_types_issue
>>
>
mpi_f08_types.mod -- to
fail, thereby causing the overall Open MPI compilation to fail.
Weird.
> On Feb 12, 2021, at 10:37 AM, Jeff Squyres (jsquyres) via users
> wrote:
>
> Can you please send all the information listed here:
>
> https://www.open-mpi.org/community/help/
>
&
Can you please send all the information listed here:
https://www.open-mpi.org/community/help/
Thanks!
On Feb 12, 2021, at 7:39 AM, Yonas Mersha via users
mailto:users@lists.open-mpi.org>> wrote:
Hi everyone,
I am trying to install openmpi-4.1.0 with the FORTRAN support for my project
On Feb 11, 2021, at 11:55 AM, Max R. Dechantsreiter
wrote:
>
> Not sure about the latest, but I built v4.1.x-202102090356-380ac96
> without errors, then used that to successfully build andd test
> GROMACS parallel mdrun.
Great!
> While I do not like using non-release versions, this is
You might want to ask your question to the Gromacs community -- they can
probably be more helpful than we can (we know little / nothing about Gromacs).
Good luck!
On Feb 11, 2021, at 10:46 AM, Wenhao Yao via users
mailto:users@lists.open-mpi.org>> wrote:
Thanks for helping me though it a
ht.us
>> # Hello from thread 1 out of 2 from process 0 out of 2 on
>> server.clearlight.us
>> # Hello from thread 0 out of 2 from process 1 out of 2 on
>> server.clearlight.us
>> # Hello from thread 1 out of 2 from process 1 out of 2 on
>> server.clearlight.us
I think Max did try the latest 4.1 nightly build (from an off-list email), and
his problem still persisted.
Max: can you describe exactly how Open MPI failed? All you said was:
>> Consequently AVX512 intrinsic functions were erroneously
>> deployed, resulting in OpenMPI failure.
Can you
Carl --
I'm not able to reproduce your error. What does config.log say?
On Feb 5, 2021, at 2:36 PM, Carl Ponder via users
mailto:users@lists.open-mpi.org>> wrote:
I usually build OpenMPI using these flags
--enable-mpi-cxx --enable-mpi-cxx-seek --enable-cxx-exceptions
but get this error for
Actually, can you try with the latest v4.1.x nightly snapshot tarball?
We have fixed several AVX-related issues since v4.1.0 was released:
https://www.open-mpi.org/nightly/v4.1.x/
On Feb 5, 2021, at 2:54 PM, George Bosilca via users
mailto:users@lists.open-mpi.org>> wrote:
Carl,
AVX
Do we know if this was definitely fixed in v4.1.x?
> On Feb 4, 2021, at 7:46 AM, Gilles Gouaillardet via users
> wrote:
>
> Martin,
>
> this is a connectivity issue reported by the btl/tcp component.
>
> You can try restricting the IP interface to a subnet known to work
> (and with no
On Jan 26, 2021, at 3:03 PM, Craig via users
mailto:users@lists.open-mpi.org>> wrote:
Is there a table somewhere that tells me what version of things like libfabric
and UCX
(and maybe compiler versions if there are known issues) are known to be good
with
which versions of OpenMPI? I poked
he configure execution, but when the system tries to
compile orte-info
it crashes. I enclose the outputs again. configure.out contains both the
outputs from configure
and compilation.
Kind regards,
Álvaro
El vie, 22 ene 2021 a las 16:09, Jeff Squyres (jsquyres)
(mailto:jsquy...@cisco.com>>) esc
Is your app calling MPI_Abort directly? There's a 2nd argument to MPI_ABORT
that should be passed to the output message. If it's not, we should
investigate that.
Or is your app aborting in some other, indirect method? If so, perhaps somehow
that 2nd argument is getting dropped somewhere
It's somewhat hard to say without more information.
What is your app doing when it calls abort?
On Jan 29, 2021, at 8:49 PM, Arturo Fernandez via users
mailto:users@lists.open-mpi.org>> wrote:
Hello,
My system is running CentOS8 & OpenMPI v4.1.0. Most stuff is working fine but
one app is
On Jan 22, 2021, at 9:49 AM, Alvaro Payero Pinto via users
wrote:
>
> I am trying to install Open MPI with Intel compiler suite for the Fortran
> side and GNU compiler suite for the C side. For factors that don’t depend
> upon me, I’m not allowed to change the C compiler suite to Intel one
Good question. I've filed https://github.com/open-mpi/ompi/issues/8379 so that
we can track this.
> On Jan 14, 2021, at 7:53 AM, Dave Love via users
> wrote:
>
> Why does 4.1 still not use the right defaults with UCX?
>
> Without specifying osc=ucx, IMB-RMA crashes like 4.0.5. I haven't
>
Yes, opening an issue would be great -- thanks!
On Dec 14, 2020, at 11:32 AM, Patrick Bégou via users
mailto:users@lists.open-mpi.org>> wrote:
OK, Thanks Gilles.
Does it still require that I open an issue for tracking ?
Patrick
Le 14/12/2020 à 14:56, Gilles Gouaillardet via users a écrit :
On Dec 12, 2020, at 4:58 AM, Lesiano 16 via users
wrote:
>
> My question is, can I assume that when skipping the beginning of the file
> that MPI will fill up with zeros? Or is it implementation dependent?
>
> I have read the standard, but I could not found anything meaningful expected
>
I have not personally tried Open MPI in WSL / WSL2.
Two suggestions:
1. Can you post the output of the error that occurs?
2. Can you try upgrading to the latest 4.0.x release, or at least the latest
v3.1.x release (as of this writing v3.1.6)?
> On Nov 27, 2020, at 10:50 AM, schoon via users
Paul --
The help message describes 4 ways to set the number of available slots on your
machine:
1. Hostfile, via "slots=N" clauses (N defaults to number of
processor cores if not provided)
2. The --host command line parameter, via a ":N" suffix on the
hostname (N defaults to 1 if not
Over the past month or so, you may have noticed that sometimes mails you sent
to the Open MPI mailing lists were delayed, sometimes by multiple days.
Our mailing list provider has been experiencing technical difficulties in
keeping the Open MPI mailing lists flowing. They tell us that they had
There's huge differences between Open MPI v2.1.1 and v4.0.3 (i.e., years of
development effort); it would be very hard to categorize them all; sorry!
What happens if you
mpirun -np 1 touch /tmp/foo
(Yes, you can run non-MPI apps through mpirun)
Is /tmp/foo created? (i.e., did the job
FYI: We've been having some problems with our mailing list provider over the
past few weeks.
No mails have been lost, but sometimes mails queue up endlessly at our mailing
list provider until a human IT staffer goes in, fixes a problem, and
effectively releases all the mails that have queued
On Oct 15, 2020, at 3:27 AM, Diego Zuccato wrote:
>
>>> The version is 3.1.3 , as packaged in Debian Buster.
>> The 3.1.x series is pretty old. If you want to stay in the 3.1.x
>> series, you might try upgrading to the latest -- 3.1.6. That has a
>> bunch of bug fixes compared to v3.1.3.
> I'm
On Oct 14, 2020, at 3:07 AM, Diego Zuccato
mailto:diego.zucc...@unibo.it>> wrote:
Il 13/10/20 16:33, Jeff Squyres (jsquyres) ha scritto:
That's odd. What version of Open MPI are you using?
The version is 3.1.3 , as packaged in Debian Buster.
The 3.1.x series is pretty old. If yo
On Oct 13, 2020, at 10:43 AM, Gus Correa via users
wrote:
>
> Can you use taskid after MPI_Finalize?
Yes. It's a variable, just like any other.
> Isn't it undefined/deallocated at that point?
No. MPI filled it in during MPI_Comm_rank() and then never touched it again.
So even though MPI
That's odd. What version of Open MPI are you using?
> On Oct 13, 2020, at 6:34 AM, Diego Zuccato via users
> wrote:
>
> Hello all.
>
> I have a problem on a server: launching a job with mpirun fails if I
> request all 32 CPUs (threads, since HT is enabled) but succeeds if I
> only request
3.1.2 was a long time ago, but I'm pretty sure that Open MPI v3.1.2 has
btl_tcp_if_include / btl_tcp_if_exclude.
Try running: "ompi_info --all --parsable | grep btl_tcp_if_"
I believe that those options will both take a CIDR notation of which network(s)
to use/not use. Note: the _if_include
On Aug 24, 2020, at 9:44 PM, Tony Ladd wrote:
>
> I appreciate your help (and John's as well). At this point I don't think is
> an OMPI problem - my mistake. I think the communication with RDMA is somehow
> disabled (perhaps its the verbs layer - I am not very knowledgeable with
> this). It
fail.
>>
>>The only good (as best I can tell) diagnostic is from openMPI.
>>ibv_obj
>>(from v2.x) complains that openib returns a NULL object, whereas
>>on my
>>server it returns logical_index=1. Can we not try to diagnose the
>>proble
Tony --
Have you tried compiling Open MPI with UCX support? This is Mellanox
(NVIDIA's) preferred mechanism for InfiniBand support these days -- the openib
BTL is legacy.
You can run: mpirun --mca pml ucx ...
> On Aug 19, 2020, at 12:46 PM, Tony Ladd via users
> wrote:
>
> One other
On Aug 7, 2020, at 12:52 PM, Oddo Da via users
mailto:users@lists.open-mpi.org>> wrote:
The Java bindings support "recent" JDK, and if you face an issue, please report
a bug (either here or on github)
Starting with Java 8, the language went into a much different direction -
functional
FWIW, we didn't talk too much about the internals of Open MPI -- but it's a
good place to start (i.e., you won't understand the internals until you
understand the externals).
You can find all the videos and slides for all 3 parts here:
https://www.open-mpi.org/video/?category=general
In
checking whether ucp_tag_send_sync_nbx is declared... no
checking whether ucp_tag_recv_nbx is declared... no
checking for ucp_request_param_t... no
configure: error: UCX support requested but not found. Aborting
.. Lana (lana.de...@gmail.com<mailto:lana.de...@gmail.com>)
On Mon, Jul
network transports automatically based on what's available.
I'll also look at the slides and see if I can make sense of them. Thanks.
.. Lana (lana.de...@gmail.com<mailto:lana.de...@gmail.com>)
On Sat, Jul 18, 2020 at 9:41 AM Jeff Squyres (jsquyres)
mailto:jsquy...@cisco.com>> wrot
On Jul 16, 2020, at 2:56 PM, Lana Deere via users
mailto:users@lists.open-mpi.org>> wrote:
I am new to open mpi. I built 4.0.4 on a CentOS7 machine and tried doing an
mpirun of a small program compiled against openmpi. It seems to have failed
because my host does not have infiniband. I
Woo hoo! I love getting emails like this. We actually spend quite a bit of
time in the design and implementation of the configure/build system so that it
will "just work" in a wide variety of situations.
Thanks!
On Jul 17, 2020, at 5:43 PM, John Duffy via users
In addition to what Gilles said, if you're using TCP for your MPI transport,
changing the MTU probably won't have a huge impact on HPL.
Open MPI will automatically react to the MTU size; there shouldn't be anything
you need to change. Indeed, when using TCP, the kernel TCP stack is the one
The slides and videos for parts 1 and 2 of the online seminar presentation "The
ABCs of Open MPI" have been posted on both the Open MPI web site and the
EasyBuild wiki:
https://www.open-mpi.org/video/?category=general
Greetings Prentice.
This is a very generic error, it's basically just indicating "somewhere in the
program, we got a bad pointer address."
It's very difficult to know if this issue is in Open MPI or in the application
itself (e.g., memory corruption by the application eventually lead to bad
ild-Tech-Talks-I:-Open-MPI
Anyone is free to join either / both parts.
Hope to see you this Wednesday!
On Jun 14, 2020, at 2:05 PM, Jeff Squyres (jsquyres) via announce
mailto:annou...@lists.open-mpi.org>> wrote:
In conjunction with the EasyBuild community, Ralph Castain (Intel, Open
is
Attached files are stdout and also config.log
Thank you!
Sahir
--
On 27. Jun 2020, at 17:48, Jeff Squyres (jsquyres)
mailto:jsquy...@cisco.com>> wrote:
On Jun 26, 2020, at 11:32 AM, Sahir Butt via users
mailto:users@lists.open-mpi.org>> wrote:
I am trying to build openmpi-4.
On Jun 26, 2020, at 7:30 AM, Peter Kjellström via users
wrote:
>
>> The cluster hardware is QLogic infiniband with Intel CPUs. My
>> understanding is that we should be using the old PSM for networking.
>>
>> Any thoughts what might be going wrong with the build?
>
> Yes only PSM will
On Jun 26, 2020, at 11:32 AM, Sahir Butt via users
wrote:
>
> I am trying to build openmpi-4.0.3 with gcc 11 on macOS. I ran following to
> configure:
>
> ../configure --prefix=/path-to/opt/openmpi-4.0.3
> --with-wrapper-ldflags="-Wl,-search_paths_first”
>
> I keep getting following error:
akes it very difficult to
> specify IP address for btl_tcp_if_include.
>
> For the named exclude interfaces, it still hanged (with no output) when I
> specified btl_base_verbose 100.
>
> I will try using the CIDR for the below hosts as an experiment.
>
> Regards,
> Vipul
>
https://www.open-mpi.org/faq/?category=tcp#ip-virtual-ip-interfaces is
referring to interfaces like "eth0:0", where the Linux kernel will have the
same index for both "eth0" and "eth0:0". This will cause Open MPI to get
confused (because it identifies Ethernet interfaces by their kernel
!
> On Jun 14, 2020, at 2:05 PM, Jeff Squyres (jsquyres) via announce
> wrote:
>
> In conjunction with the EasyBuild community, Ralph Castain (Intel, Open MPI,
> PMIx) and Jeff Squyres (Cisco, Open MPI) will host an online presentation
> about Open MPI on **Wednesday June 24th 2020
On Jun 19, 2020, at 6:59 PM, Thomas M. Payerle via users
wrote:
>
> We are upgrading a cluster from RHEL6 to RHEL8, and have migrated some nodes
> to a new partition and reimaged with RHEL8. I am having some issues getting
> openmpi to work with infiniband on the nodes upgraded to RHEL8.
On Jun 15, 2020, at 3:43 PM, Roberto Herraro via users
wrote:
>
> We have a small cluster and are running paired HPL to test performance, but
> are getting poor results. One of our suspicions is that the regular 1GbE
> interface might be being used, rather than the 100G interface. Is there a
Martin --
Someone filed the same issue shortly before your post:
https://github.com/open-mpi/ompi/issues/7810
On Jun 12, 2020, at 6:10 PM, Martin Audet via users
mailto:users@lists.open-mpi.org>> wrote:
Hello OMPI_Developers,
When I compile my C++ code with Open MPI version 4.0.3 or
In conjunction with the EasyBuild community, Ralph Castain (Intel, Open MPI,
PMIx) and Jeff Squyres (Cisco, Open MPI) will host an online presentation about
Open MPI on **Wednesday June 24th 2020** at:
- 11am US Eastern time
- 8am US Pacific time
- 3pm UTC
- 5pm CEST
The general scope of the
You cited Open MPI v2.1.1. That's a pretty ancient version of Open MPI.
Any chance you can upgrade to Open MPI 4.0.x?
> On Jun 5, 2020, at 7:24 PM, Stephen Siegel wrote:
>
>
>
>> On Jun 5, 2020, at 6:55 PM, Jeff Squyres (jsquyres)
>> wrote:
>>
>>
On Jun 5, 2020, at 6:35 PM, Stephen Siegel via users
wrote:
>
> [ilyich:12946] 3 more processes have sent help message help-mpi-btl-base.txt
> / btl:no-nics
> [ilyich:12946] Set MCA parameter "orte_base_help_aggregate" to 0 to see all
> help / error messages
It looks like your output somehow
Are you actually running into a problem?
A successful install may still end with "Nothing to be done..." messages.
On Jun 5, 2020, at 10:48 AM, Edris Tajfirouzeh via users
mailto:users@lists.open-mpi.org>> wrote:
Dear Operator
I'm trying to install open-mpi package on my mac catalina version
On top of what Ralph said, I think that this output is unexpected:
> Starting server daemon at host "cod5"Starting server daemon at host
> "cod6"Starting server daemon at host "has4"Starting server daemon at host "co
> d4"
>
>
>
> Starting server daemon at host "hpb12"Starting server daemon
Was your Open MPI compiled with -r8?
We definitely recommend using the same compiler flags to compile Open MPI as
your application. As George noted, -r8 can definitely cause issues if one was
compiled with it and the other was not.
On May 18, 2020, at 12:24 AM, George Bosilca via users
On May 12, 2020, at 7:42 AM, Leandro wrote:
>
> I compile it statically to make sure compilers libraries will not be a
> dependency, and I do this way for years.
For what it's worth, you're compiling with -static-intel, which should take
care of removing the compiler's libraries as
It looks like you are building both static and dynamic libraries
(--enable-static and --enable-shared). This might be confusing the issue -- I
can see at least one warning:
icc: warning #10237: -lcilkrts linked in dynamically, static library not
available
It's not easy to tell from the
On Apr 23, 2020, at 8:50 AM, Patrick Bégou
wrote:
>
> As we say in french "dans le mille!" you were right.
> I'm not the admin of these servers and a "mpirun not found" was sufficient in
> my mind. It wasn't.
>
> As I had deployed OpenMPI 4.0.2 I launch a new build after setting my
>
The test should only take a few moments; no need to let it sit for a full hour.
I have seen this kind of behavior before if you have an Open MPI installation
in your PATH / LD_LIBRARY_PATH already, and then you invoke "make check".
Because the libraries may be the same name and/or .so version
Can you send all the information listed here:
https://www.open-mpi.org/community/help/
On Apr 15, 2020, at 6:13 AM, フォンスポール Ｊ via users
mailto:users@lists.open-mpi.org>> wrote:
Dear Gilles,
Thank you for your advice. I have tried a few of the suggestions that I
encountered following
On Mar 26, 2020, at 5:36 AM, Raut, S Biplab wrote:
>
> I am doing pairwise send-recv and not all-to-all since not all the data is
> required by all the ranks.
> And I am doing blocking send and recv calls since there are multiple
> iterations of such message chunks to be sent with
On Mar 25, 2020, at 4:49 AM, Raut, S Biplab via users
wrote:
>
> Let’s say the application is running with 128 ranks.
> Each rank is doing send() msg to rest of 127 ranks where the msg length sent
> is under question.
> Now after all the sends are completed, each rank will recv() msg from
Let's back up and ask a question: is there a reason you're using Bsend?
I.e., do you need to use Bsend for some reason, or can you use regular
(potentially non-buffering) sends instead?
On Mar 18, 2020, at 5:16 AM, Martyn Foster via users
mailto:users@lists.open-mpi.org>> wrote:
Hi George,
> On Mar 13, 2020, at 9:33 AM, Jeffrey Layton via users
> wrote:
>
> Good morning,
>
> I've compiled a hello world MPI code and when I run it, I get some messages
> I'm not familiar with. The first one is,
>
> --
>
If you're using Open MPI v4.0.x, you should likely be using the UCX PML plugin
for InfiniBand communication.
As I understand it, UCX is controlled by environment variables. You'll likely
need to look through the UCX documentation to see what environment variable(s)
is(are) needed for setting
> On Feb 3, 2020, at 12:35 PM, Bennet Fauber wrote:
>
> This is what CentOS installed.
>
> $ yum list installed hwloc\*
> Loaded plugins: langpacks
> Installed Packages
> hwloc.x86_64 1.11.8-4.el7
> @os
> hwloc-devel.x86_64
On Feb 3, 2020, at 10:03 AM, Bennet Fauber wrote:
>
> Ah, ha!
>
> Yes, that seems to be it. Thanks.
Ok, good. I understand that UCX is the "preferred" mechanism for IB these days.
> If I might, on a configure related note ask, whether, if we have
> these installed with the CentOS 7.6 we
Bennet --
Just curious: is there a reason you're not using UCX?
> On Feb 2, 2020, at 4:06 PM, Bennet Fauber via users
> wrote:
>
> We get these warnings/error from OpenMPI, version 3.1.4 and 4.0.2
>
> --
> WARNING: No
Can you please send all the information listed here:
https://www.open-mpi.org/community/help/
Thanks!
On Jan 27, 2020, at 12:00 PM, Collin Strassburger via users
mailto:users@lists.open-mpi.org>> wrote:
Hello,
I had initially thought the same thing about the streams, but I have 2
Greetings Mehmet.
I'm curious: why do you want to use static linking? It tends to cause
complications and issues like this.
Specifically: Open MPI's `--disable-shared --enable-static` switches means that
Open MPI will produce static libraries instead of shared libraries (e.g.,
libmpi.a
In addition to that, let me pile on to what Michael Heinz said (he suggested
you use "... --mca btl_base_verbose 100 ..." In your mpirun command line).
Short version:
Try "mpirun --mca pml_base_verbose 100 --mca btl_base_verbose 100 ..." I.e.,
add in the bit about pml_base_verbose.
If you see
On Jan 7, 2020, at 11:48 AM, Pablo Goloboff via users
mailto:users@lists.open-mpi.org>> wrote:
Hi,
I am trying to port an existing PVM-based application (developed over many
years) onto MPI, trying to preserve as much of the organization as possible.
This is my first contact with MPI, after
> On Dec 13, 2019, at 10:08 AM, Tom Rosmond via users
> wrote:
>
> I recently upgraded to Fedora 31 and tried to build openmpi-4.0.2 configured
> with
>
> ./configure FC=gfortran F90=gfortran --prefix=/opt/openmpi --with-slurm
>
> The configure seemed fine, but the 'make' failed with
>
>
On Dec 6, 2019, at 1:03 PM, Jeff Squyres (jsquyres) via users
wrote:
>
>> I get the same error when running in a single node. I will try to use the
>> last version. Is there way to check if different versions of open mpi were
>> used in different nodes?
>
> mp
On Dec 6, 2019, at 12:40 PM, Guido granda muñoz wrote:
>
> I get the same error when running in a single node. I will try to use the
> last version. Is there way to check if different versions of open mpi were
> used in different nodes?
mpirun -np 2 ompi_info | head
Or something like that.
101 - 200 of 1536 matches
Mail list logo
