[Mpi-forum] MPI Next leadership

Anthony Skjellum tony at cis.uab.edu
Thu Sep 27 07:18:11 CDT 2012


Besides leadership, we should ask "why" MPI Next "now" vs "later" vs. "ever."  

According to Robert's Rules of Order, a motion to adjourn is always
in order.  Why are we assuming that MPI goes on as before, with multiyear planning?
We should first entertain, second, and either accept or defeat a motion to adjourn, 
and those voting against adjourning should provide rationale for continuing "now."  
A second vote could be to "wait X months/years" and then automatically return.

For instance, why not give the OpenMPI and MPICH development teams a chance to finish their
work, distribute releases, write up findings, discover errata, tune, let people get trained,
let applications be ported and use those new features, achieve buy-in and acceptance, and have 
actual users give feedback, so we come tothe "Next" with the experience of results?

In the meantime, research on the areas that are still controversial could continue without
the need to "be standardized" at the same time as they are being figured out.  




----- Original Message -----
From: "Rolf Rabenseifner" <rabenseifner at hlrs.de>
To: "Main MPI Forum mailing list" <mpi-forum at lists.mpi-forum.org>
Sent: Thursday, September 27, 2012 5:17:15 AM
Subject: Re: [Mpi-forum] MPI Next leadership

As MPI-2.1 chair, I must clearly say that I could not have
been the MPI-2.1 chair if there would not have been Rich as 
the convener of the meetings.
And I'm thankful that Rich has done it the whole time together
with the local organizers, mainly Jeff, the team of Microsoft
and all European organizers at the EuroMPI conferences.

As far as I understand, Rich would continue to be our convener.

About MPI-next chair:
Although the chair should clearly be neutral, I would prefer
to have a MPI-next chair from a public research institution.
Of course he/she must have the support from her/his institution
to travel to the meetings.
I see also that he/she need not only to focus on the new
topics, but also on the quality of the standard, i.e., after 
the time and content is right, to organizing also the finalization 
of such an new version of MPI.

Rolf

----- Original Message -----
> From: "Rajeev Thakur" <thakur at mcs.anl.gov>
> To: "Main MPI Forum mailing list" <mpi-forum at lists.mpi-forum.org>
> Sent: Thursday, September 27, 2012 10:51:45 AM
> Subject: Re: [Mpi-forum] MPI Next leadership
> It has also been suggested that the Forum take a break for a bit
> before resuming again, but I don't know if that was discussed at the
> last meeting.
> 
> Rajeev
> 
> 
> 
> On Sep 27, 2012, at 1:09 AM, Schulz, Martin wrote:
> 
> > On Sep 26, 2012, at 9:46 PM, Bronis R. de Supinski wrote:
> >
> >>
> >> Jeff:
> >>
> >> My intent is not to "neuter" the chair. Instead, I am only
> >> stating that the chair should clearly state when (s)he is
> >> voicing personal/institutional opinions as opposed to acting
> >> as leader (sometimes the roles conflict and also sometimes
> >> the leader does have to enforce the clear general opinion).
> >
> > I think you are saying the same thing. This is, of course, aside the
> > fact that you wouldn't find anybody who is active in the forum and
> > who doesn't have a personal agenda. We all have one and that's why
> > we are engaged. The people with a double role just have to keep it
> > separate - and that's not only true for the chair, but (perhaps to a
> > lesser extent) also for the other roles that Jeff pointed out.
> >
> > I also agree with Jeff - if we are setting up a new structure, let's
> > look at all roles and define them better.
> >
> >> As to logistics, yes, I see your point but I still think
> >> that combining them would generally simplify things. I
> >> also think it is fine for the chair to delegate aspects
> >> that the convener handles but they should still be
> >> responsible for them (we all blame the chair if they do
> >> not go right so they should also have the authority since
> >> they de facto have responsibility).
> >
> > The convener role will always have to be split up, in particular if
> > meetings get hosted at individual organizations. A certain part of
> > the work (e.g., wireless, badging, room reservation, ...) always
> > fell on people from the hosting organization and that won't change.
> >
> > The more critical question, IMHO, is the financial part that Jeff
> > brought up. We have organizations that have more restrictions than
> > others and we have to take that into account. Further, if the person
> > responsible for financial transactions changes, we have to transfer
> > money - the same if the person responsible changes organization (as
> > we can see at Rich's example, which ended up being non trivial or
> > even impossible, if I understood the situation right). The easiest
> > thing would be to have a constant entity who manages all financial
> > aspects of the forum along with a separate role of a treasurer.
> >
> > Martin
> >
> >
> >>
> >> Bronis
> >>
> >>
> >> On Wed, 26 Sep 2012, Jeff Squyres wrote:
> >>
> >>> This is exactly why we should have the discussion: "chair" and
> >>> "convener" have not necessarily been the same person over the last
> >>> few years. Rich has been the convener since 2.1, but has only been
> >>> the chair for 3.0.
> >>>
> >>> My $0.02 is that I don't think they need to be the same person.
> >>> The chair is the [neutral] technical leader (and therefore sets
> >>> the meeting agendas), but the convener is someone with an
> >>> organization who can make meeting room reservations, accept
> >>> registration and payments from meeting attendees, etc. That being
> >>> said, as long as the chair remains neutral in setting the agenda
> >>> and guiding the process, the chair should be allowed to have their
> >>> own technical opinions, just like anyone else. Otherwise, you're
> >>> just neutering the chair, and their incentive to participate is
> >>> quite limited.
> >>>
> >>> In short: I see the convener as a largely logistical role, but the
> >>> chair as more of a technical role.
> >>>
> >>>
> >>>
> >>> On Sep 27, 2012, at 6:33 AM, Bronis R. de Supinski wrote:
> >>>
> >>>>
> >>>> Personally, I think "chair" should also be "meeting convener".
> >>>> The chair should be largely administrative/neutral/meeting
> >>>> organizer and runner. Convening meetings is a natural part
> >>>> of that activity. The chair should not be promoting a
> >>>> particular technical agenda (at least not as "chair").
> >>>>
> >>>>
> >>>> On Wed, 26 Sep 2012, Jeff Squyres wrote:
> >>>>
> >>>>> As part of this discussion, it might be worthwhile to also bring
> >>>>> up the auxiliary roles -- there's a bunch of things that happen
> >>>>> behind the scenes to make the Forum function:
> >>>>>
> >>>>> - secretary
> >>>>> - meeting convener
> >>>>> - web maintainers (main site, meetings site, lists site)
> >>>>> - svn / trac maintainer
> >>>>>
> >>>>> I mention this because, although I'm happy to continue in the
> >>>>> role as secretary, a) if there's someone else burning to do that
> >>>>> role, it would be fair to discuss it, and b) sometimes there has
> >>>>> been confusion about exactly what the secretary is supposed to
> >>>>> do. Maybe it would be useful to delineate the responsibilities
> >>>>> of the auxiliary roles. Just like we discovered that we all
> >>>>> didn't have the same conceptions of how the secretary applies
> >>>>> the voting rules, perhaps there are other unwritten
> >>>>> inconsistencies, too.
> >>>>>
> >>>>>
> >>>>>
> >>>>> On Sep 26, 2012, at 10:27 PM, Brightwell, Ronald wrote:
> >>>>>
> >>>>>>
> >>>>>> Greetings:
> >>>>>>
> >>>>>> I wasn't able to attend the Forum meeting last week, so I don't
> >>>>>> know if this came up or not, but given that it's election
> >>>>>> season in the US and MPI 3.0 has been officially accepted, it
> >>>>>> seems like a good time to discuss leadership of the Forum going
> >>>>>> forward. I know several people who have been involved from a
> >>>>>> leadership standpoint are in different situations now then they
> >>>>>> were when the 3.0 process started, and their priorities and
> >>>>>> commitments are different. I think it might be good to have an
> >>>>>> open discussion about making changes to some of the leadership
> >>>>>> roles in an effort to make sure that we start this next phase
> >>>>>> in the best way possible.
> >>>>>>
> >>>>>> -Ron
> >>>>>>
> >>>>>>
> >>>>>>
> >>>>>> _______________________________________________
> >>>>>> mpi-forum mailing list
> >>>>>> mpi-forum at lists.mpi-forum.org
> >>>>>> http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi-forum
> >>>>>
> >>>>>
> >>>>> --
> >>>>> Jeff Squyres
> >>>>> jsquyres at cisco.com
> >>>>> For corporate legal information go to:
> >>>>> http://www.cisco.com/web/about/doing_business/legal/cri/
> >>>>>
> >>>>>
> >>>>> _______________________________________________
> >>>>> mpi-forum mailing list
> >>>>> mpi-forum at lists.mpi-forum.org
> >>>>> http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi-forum
> >>>>>
> >>>> _______________________________________________
> >>>> mpi-forum mailing list
> >>>> mpi-forum at lists.mpi-forum.org
> >>>> http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi-forum
> >>>
> >>>
> >>> --
> >>> Jeff Squyres
> >>> jsquyres at cisco.com
> >>> For corporate legal information go to:
> >>> http://www.cisco.com/web/about/doing_business/legal/cri/
> >>>
> >>>
> >> _______________________________________________
> >> mpi-forum mailing list
> >> mpi-forum at lists.mpi-forum.org
> >> http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi-forum
> >
> > ________________________________________________________________________
> > Martin Schulz, schulzm at llnl.gov, http://people.llnl.gov/schulzm
> > CASC @ Lawrence Livermore National Laboratory, Livermore, USA
> >
> >
> >
> >
> > _______________________________________________
> > mpi-forum mailing list
> > mpi-forum at lists.mpi-forum.org
> > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi-forum
> 
> 
> _______________________________________________
> mpi-forum mailing list
> mpi-forum at lists.mpi-forum.org
> http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi-forum

-- 
Dr. Rolf Rabenseifner . . . . . . . . . .. email rabenseifner at hlrs.de
High Performance Computing Center (HLRS) . phone ++49(0)711/685-65530
University of Stuttgart . . . . . . . . .. fax ++49(0)711 / 685-65832
Head of Dpmt Parallel Computing . . . www.hlrs.de/people/rabenseifner
Nobelstr. 19, D-70550 Stuttgart, Germany . (Office: Allmandring 30)
_______________________________________________
mpi-forum mailing list
mpi-forum at lists.mpi-forum.org
http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi-forum

-- 
Anthony Skjellum, PhD
Professor and Chair
Dept. of Computer and Information Sciences
Director, UAB Center for Information Assurance and Joint Forensics Research ("The Center")
University of Alabama at Birmingham
+1-(205)934-8657; FAX: +1- (205)934-5473

___________________________________________
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.

Please consider the environment before printing this e-mail 



More information about the mpi-forum mailing list