[mpiwg-ft] FTWG Call Topics?

Ignacio Laguna lagunaperalt1 at llnl.gov
Wed Jun 21 22:47:43 CDT 2017


That would be hard I think. We would need to redefine what MPI_Init is 
(e.g., make it optional). I have to admit we have not considered this 
case carefully though.


-- 
Ignacio Laguna
Center for Applied Scientific Computing (CASC)
Lawrence Livermore National Laboratory
Phone: 925-422-7308, Fax: 925-422-6287

On 6/21/17 2:07 PM, Anthony Skjellum wrote:
> Hi, Could we write MPI programs with only MPI_Reinit()?  Skip MPI_Init()...
> Regards,
> Tony
> 
> 
> On Wed, Jun 21, 2017 at 2:44 PM, Aurelien Bouteiller 
> <bouteill at icl.utk.edu <mailto:bouteill at icl.utk.edu>> wrote:
> 
>     The minutes of the call discussing the Init issue can be found here:
>     https://github.com/mpiwg-ft/ft-issues/wiki/2017-06-21
>     <https://github.com/mpiwg-ft/ft-issues/wiki/2017-06-21>
> 
>>     On Jun 21, 2017, at 13:54, Bland, Wesley <wesley.bland at intel.com
>>     <mailto:wesley.bland at intel.com>> wrote:
>>
>>     Sounds good. We'll go ahead and hold the call then.
>>
>>>     On Jun 21, 2017, at 12:41 PM, Aurelien Bouteiller
>>>     <bouteill at icl.utk.edu <mailto:bouteill at icl.utk.edu>> wrote:
>>>
>>>     I am available today for discussion. I think we can use some time
>>>     to discuss the MPI Init issue, and keep the FA-MPI presentation
>>>     for next meeting if that’s ok with all.
>>>
>>>     We can use the Zoom room for the meeting today at 3pm EST
>>>     :
>>>     Aurelien Bouteiller is inviting you to a scheduled Zoom meeting.
>>>
>>>     Topic: MPI Forum FT-WG
>>>     Time: this is a recurring meeting Meet anytime
>>>
>>>     Join from PC, Mac, Linux, iOS or Android:
>>>     https://tennessee.zoom.us/j/632356722?pwd=lI4%2F169CGcewIumekTziMw%3D%3D
>>>     <https://tennessee.zoom.us/j/632356722?pwd=lI4%2F169CGcewIumekTziMw%3D%3D>
>>>       Password: mpiforum
>>>
>>>     Or iPhone one-tap (US Toll): +14086380968
>>>     <tel:(408)%20638-0968>,632356722# or +16465588656
>>>     <tel:(646)%20558-8656>,632356722#
>>>
>>>     Or Telephone:
>>>       Dial: +1 408 638 0968 <tel:(408)%20638-0968> (US Toll) or +1
>>>     646 558 8656 <tel:(646)%20558-8656> (US Toll)
>>>       Meeting ID: 632 356 722
>>>       International numbers available:
>>>     https://tennessee.zoom.us/zoomconference?m=GscM59o_Qoig8v4aJl1OrsnXL-7Blrke
>>>     <https://tennessee.zoom.us/zoomconference?m=GscM59o_Qoig8v4aJl1OrsnXL-7Blrke>
>>>
>>>     Or an H.323/SIP room system:
>>>       H.323: 162.255.37.11 (US West) or 162.255.36.11 (US East)
>>>       Meeting ID: 632 356 722
>>>       Password: 366244
>>>
>>>       SIP: 632356722 at zoomcrc.com <mailto:632356722 at zoomcrc.com>
>>>       Password: 366244
>>>
>>>
>>>     Aurelien
>>>
>>>>     On Jun 21, 2017, at 13:29, Anthony Skjellum <skjellum at auburn.edu
>>>>     <mailto:skjellum at auburn.edu>> wrote:
>>>>
>>>>     Hi all, I am OK if we cancel too, but at the next meeting we
>>>>     hold, I wanted to do an update about FA-MPI work and answer
>>>>     questions people have about our approach to it.
>>>>     Since this came up a couple calls ago.  It would be fine to wait
>>>>     for next call slot to do this... We are actively extending
>>>>     FA-MPI at this time.
>>>>     Tony
>>>>
>>>>
>>>>     Anthony Skjellum, PhD
>>>>     Professor of Computer Science and Software Engineering and
>>>>      Charles D. McCrary Eminent Scholar Endowed Chair
>>>>     Director of the Charles D. McCrary Institute
>>>>     Samuel Ginn College of Engineering
>>>>     Auburn University
>>>>     e-mail: skjellum at auburn.edu <mailto:skjellum at auburn.edu> or
>>>>     skjellum at gmail.com <mailto:skjellum at gmail.com>
>>>>     web sites: http://cyber.auburn.edu http://mccrary.auburn.edu
>>>>     cell: +1-205-807-4968 <tel:(205)%20807-4968> ; office:
>>>>     +1-334-844-6360 <tel:(334)%20844-6360>
>>>>
>>>>     CONFIDENTIALITY: This e-mail and any attachments are
>>>>     confidential and
>>>>     may be privileged. If you are not a named recipient, please
>>>>     notify the
>>>>     sender immediately and do not disclose the contents to another
>>>>     person,
>>>>     use it for any purpose or store or copy the information in any
>>>>     medium.
>>>>
>>>>     ________________________________________
>>>>     From: mpiwg-ft <mpiwg-ft-bounces at lists.mpi-forum.org
>>>>     <mailto:mpiwg-ft-bounces at lists.mpi-forum.org>> on behalf of
>>>>     Ignacio Laguna <lagunaperalt1 at llnl.gov
>>>>     <mailto:lagunaperalt1 at llnl.gov>>
>>>>     Sent: Wednesday, June 21, 2017 12:17 PM
>>>>     To: MPI WG Fault Tolerance and Dynamic Process Control working
>>>>     Group; Bland, Wesley
>>>>     Subject: Re: [mpiwg-ft] FTWG Call Topics?
>>>>
>>>>     I'm okay if we cancel.
>>>>
>>>>     On a different note, I wanted to ask the group about the error
>>>>     handler
>>>>     for MPI_Init. As I look into the problem of MPI re-initialization, I
>>>>     came across some questions, which I'm not been able to answer
>>>>     myself by
>>>>     reading the standard:
>>>>
>>>>     - Since one cannot set an error handler before MPI_Init, is is
>>>>     expected
>>>>     that MPI_ERRORS_ARE_FATAL is enabled when MPI_Init is called? If
>>>>     that is
>>>>     the case, one can assume that if an error occurs the first time
>>>>     MPI_Init
>>>>     is called (let's assume in the future we can call MPI_Init several
>>>>     times), MPI will abort?
>>>>
>>>>     - Is there any situation, in which MPI_Init could return an
>>>>     error code
>>>>     if a failure occurs?
>>>>
>>>>     - Is there a way to set MPI_ERROR_RETURN in COMM_WORLD before
>>>>     calling
>>>>     MPI_Init?
>>>>
>>>>     I realize this was discussed in 2010, but I'm not sure if the forum
>>>>     found a solution:
>>>>     https://lists.mpi-forum.org/pipermail/mpiwg-ft/2010-August/000500.html
>>>>     <https://lists.mpi-forum.org/pipermail/mpiwg-ft/2010-August/000500.html>
>>>>
>>>>     Thanks!
>>>>
>>>>     --
>>>>     Ignacio Laguna
>>>>     Center for Applied Scientific Computing (CASC)
>>>>     Lawrence Livermore National Laboratory
>>>>     Phone: 925-422-7308 <tel:(925)%20422-7308>, Fax: 925-422-6287
>>>>     <tel:(925)%20422-6287>
>>>>
>>>>     On 6/21/17 9:40 AM, Bland, Wesley wrote:
>>>>>     I don't have any topics for today's FTWG con call. Does anyone
>>>>>     have anything they want to discuss?
>>>>>
>>>>>     Should we cancel?
>>>>>     _______________________________________________
>>>>>     mpiwg-ft mailing list
>>>>>     mpiwg-ft at lists.mpi-forum.org <mailto:mpiwg-ft at lists.mpi-forum.org>
>>>>>     https://lists.mpi-forum.org/mailman/listinfo/mpiwg-ft
>>>>>     <https://lists.mpi-forum.org/mailman/listinfo/mpiwg-ft>
>>>>>
>>>>     _______________________________________________
>>>>     mpiwg-ft mailing list
>>>>     mpiwg-ft at lists.mpi-forum.org <mailto:mpiwg-ft at lists.mpi-forum.org>
>>>>     https://lists.mpi-forum.org/mailman/listinfo/mpiwg-ft
>>>>     <https://lists.mpi-forum.org/mailman/listinfo/mpiwg-ft>
>>>>     _______________________________________________
>>>>     mpiwg-ft mailing list
>>>>     mpiwg-ft at lists.mpi-forum.org <mailto:mpiwg-ft at lists.mpi-forum.org>
>>>>     https://lists.mpi-forum.org/mailman/listinfo/mpiwg-ft
>>>>     <https://lists.mpi-forum.org/mailman/listinfo/mpiwg-ft>
>>>
>>
> 
> 
>     _______________________________________________
>     mpiwg-ft mailing list
>     mpiwg-ft at lists.mpi-forum.org <mailto:mpiwg-ft at lists.mpi-forum.org>
>     https://lists.mpi-forum.org/mailman/listinfo/mpiwg-ft
>     <https://lists.mpi-forum.org/mailman/listinfo/mpiwg-ft>
> 
> 
> 
> 
> -- 
> Anthony Skjellum, PhD
> skjellum at gmail.com <mailto:skjellum at gmail.com>
> Cell: +1-205-807-4968
> 
> 


More information about the mpiwg-ft mailing list