Open MPI logo

MTT Devel Mailing List Archives

  |   Home   |   Support   |   FAQ   |   all MTT Users mailing list

From: Jelena Pjesivac-Grbovic (pjesa_at_[hidden])
Date: 2007-09-20 13:55:49


I already have that - but it did not generate anything when I call it
second time (it generates output the first time when I execute mtt client
with all sections).
(I even tried adding --section reporter to the command line)

On Thu, 20 Sep 2007, Jeff Squyres wrote:

> Yes, you can add another [reporter] section to the INI file to have
> it give you a report locally. For example, something like this would
> give you a summary table at the end:
>
> [Reporter: text file]
> module = TextFile
>
> textfile_filename =
>
> # User-defined report headers/footers
> textfile_summary_header = <<EOT
> hostname: &shell("hostname")
> uname: &shell("uname -a")
> who am i: &shell("who am i")
> EOT
>
> textfile_summary_footer =
> textfile_detail_header =
> textfile_detail_footer =
>
> textfile_textwrap = 78
>
>
> On Sep 20, 2007, at 12:40 PM, Jelena Pjesivac-Grbovic wrote:
>
>> Thanks,
>> I have another question about MTT reporting. I am not sure what
>> part am I
>> missing.
>> I executed tests on 8 nodes (included MPI Get, MPI Install, and all
>> other
>> phases)
>> and then I wanted to run 36 node test using something like:
>> ./client/mtt --scratch /home/pjesa/mtt/scratch \
>> --file /home/pjesa/mtt/collective-bakeoff/samples/ompi-core-perf-
>> testing_openmpi.ini \
>> --verbose --print-time --section 'run;skampi'
>>
>> The test runs but I receive no output/report. Is there some
>> additional
>> flag to specify that I want to the output/report at the end?
>>
>> Here is the stdout of the test:
>> --------
>> ** MTT: ./client/mtt --scratch /home/pjesa/mtt/scratch --file
>> /home/pjesa/mtt/collective-bakeoff/samples/ompi-core-perf-
>> testing_openmpi
>> .ini --verbose --print-time --section run;skampi
>> *** Reporter initializing
>> *** Reporter initialized
>> *** MPI get phase starting
>> *** MPI get phase complete
>>>> Phase: MPI Get
>> Started: Thu Sep 20 11:41:42 2007
>> Stopped: Thu Sep 20 11:41:42 2007
>> Elapsed: 00:00:00
>> Total elapsed: 00:00:00
>> *** MPI install phase starting
>> *** MPI install phase complete
>>>> Phase: MPI Install
>> Started: Thu Sep 20 11:41:42 2007
>> Stopped: Thu Sep 20 11:41:42 2007
>> Elapsed: 00:00:00
>> Total elapsed: 00:00:00
>> *** Test get phase starting
>> *** Test get phase complete
>>>> Phase: Test Get
>> Started: Thu Sep 20 11:41:42 2007
>> Stopped: Thu Sep 20 11:41:42 2007
>> Elapsed: 00:00:00
>> Total elapsed: 00:00:00
>> *** Test build phase starting
>> *** Test build phase complete
>>>> Phase: Test Build
>> Started: Thu Sep 20 11:41:42 2007
>> Stopped: Thu Sep 20 11:41:42 2007
>> Elapsed: 00:00:00
>> Total elapsed: 00:00:00
>> *** Run test phase starting
>>>> Test run [skampi]
>>>> Running with [ompi-nightly-v1.2] / [1.2.4rc1r16161] /
>> [ompi/gnu-standard]
>> Using MPI Details [ompi] with MPI Install [ompi/gnu-standard]
>> Total of 1 tests to run in this section
>> Test: skampi, np=36, variant=1: Passed
>> ....
>> Test: skampi, np=36, variant=70: Passed
>> ### Test progress: 1 of 1 section tests complete (100%)
>> *** Run test phase complete
>>>> Phase: Test Run
>> Started: Thu Sep 20 11:41:42 2007
>> Stopped: Thu Sep 20 12:13:06 2007
>> Elapsed: 00:31:24
>> Total elapsed: 00:31:24
>>>> Phase: Trim
>> Started: Thu Sep 20 12:13:06 2007
>> Stopped: Thu Sep 20 12:13:06 2007
>> Elapsed: 00:00:00
>> Total elapsed: 00:31:24
>> *** Reporter finalizing
>> *** Reporter finalized
>> --------
>>
>>
>>
>> On Tue, 18 Sep 2007, Jeff Squyres wrote:
>>
>>> On Sep 18, 2007, at 6:36 PM, Jeff Squyres wrote:
>>>
>>>> Doh -- looks like a corner case that I missed (if you use a
>>>> hostfile,
>>>> it should automatically set the resource manager to "none", I would
>>>> think). Did you set a value for "resource_manager" in the INI file?
>>>> If so, take it out or just set it to none. I'll file a bug to fix
>>>> this tomorrow (i.e., if you don't set it at all and are using a
>>>> hostlist/hostfile, the Right Thing happens).
>>>
>>> Oops! I missed your statement:
>>>
>>>>> I think I should set resource manager to none - but I am not sure
>>>>> where.
>>>
>>> You can set "resource_manager = none" in the MPI Details section.
>>> But MTT should be doing this for you automatically; I'll fix it
>>> tomorrow.
>>>
>>> --
>>> Jeff Squyres
>>> Cisco Systems
>>>
>>> _______________________________________________
>>> mtt-users mailing list
>>> mtt-users_at_[hidden]
>>> http://www.open-mpi.org/mailman/listinfo.cgi/mtt-users
>>>
>>
>> --
>> Jelena Pjesivac-Grbovic, Pjesa
>> Graduate Research Assistant
>> Innovative Computing Laboratory
>> Computer Science Department, UTK
>> Claxton Complex 350
>> (865) 974 - 6722
>> (865) 974 - 6321
>> jpjesiva_at_[hidden]
>>
>> "The only difference between a problem and a solution is that
>> people understand the solution."
>> -- Charles Kettering
>>
>> _______________________________________________
>> mtt-users mailing list
>> mtt-users_at_[hidden]
>> http://www.open-mpi.org/mailman/listinfo.cgi/mtt-users
>
>
> --
> Jeff Squyres
> Cisco Systems
>
> _______________________________________________
> mtt-users mailing list
> mtt-users_at_[hidden]
> http://www.open-mpi.org/mailman/listinfo.cgi/mtt-users
>

--
Jelena Pjesivac-Grbovic, Pjesa
Graduate Research Assistant
Innovative Computing Laboratory
Computer Science Department, UTK
Claxton Complex 350
(865) 974 - 6722 
(865) 974 - 6321
jpjesiva_at_[hidden]
"The only difference between a problem and a solution is that
  people understand the solution."
                                       -- Charles Kettering