[Mpi-forum] 2011 meetings

Bronis R. de Supinski bronis at llnl.gov
Thu Jul 8 17:34:26 CDT 2010


Rolf:

Re:
> 5 meetings in San Francisco and Chicago means
> 3+2 or 2+3.

The problem with this math is that it treats each year
as a unit when the actual unit is individual meetings
or the overall process. In 2010, we have:

ATL, SFO, ORD, SFO, EURO, ORD, SFO.

You propose to follow this in 2011 with:

SFO, ORD, SFO, ORD, EURO, SFO

My claim is we then in 2012 would be led to have:

SFO, ORD, SFO, ORD, EURO, SFO

Dropping the EuroMPI meetings we then have 9 out of 15
meetings in California. That is hardly equal. More
importantly, we have had several people express that
costs make it so they can attend Chicago meetings more
easily than California (yes, many are fine with California
meetings but not all). The arguments against Chicago are
all weather and none about participation (the only thing
close is the "significantly better for Japanese attendees"
claim, which I think is a dubious: with United, I find 3
nonstops a day between SFO and Narita each way a day and
those flights take about ten hours while I find 2 non-stops
a day each way between ORD and Narita that take about 12 hours,
so we are talking about a two flight time difference; yes, I
know they are long flights but either way the cost is really
a day on each end so the time difference is negligible).

> With Chicago in winter time, the question is not the temperature,
> it is the snow which sometimes continuously falls 3 days
> (as I read).

I do not recal EVER seeing a three day snow storm in
the eight years I lived in Chicago. Yes, the weather
is better in California, I cannot argue that but I
think participation is more important.

I think the more important question to raise here is
the one Torsten was initiating: can we get the NSF
funding renewed? $750 is tight for most of these trips
but it much better than nothing (I don't use it although
we do have Martin use so I am somewhat neutral here;
I think it is essential for many of the US academics).

Bronis




> It is bad to make a 48h meeting slots
> and the airplanes have problems to arrive.
> Okay, some may have problems at their starting point
> but not the whole forum.
> Therefore, the sequence SFO - ORD - SFO - ORD - (SFO in Dec?)
> seems quite natural.
> And the Dec.-meeting should be not needed.
> We wanted to finish before SC, i.e., at EuroMPI meeting.
>
> And SFO is a bit worse for Europe, but significantly better
> for Japanese attendees.
>
> Rich, what is your opinion now?
>
> Best regards
> Rolf
>
>
> ----- Original Message -----
>> From: "Bronis R. de Supinski" <bronis at llnl.gov>
>> To: "Main MPI Forum mailing list" <mpi-forum at lists.mpi-forum.org>
>> Sent: Thursday, July 8, 2010 5:18:35 PM
>> Subject: Re: [Mpi-forum] 2011 meetings
>> Rolf:
>> 
>> Re:
>> > For me, Atlanta is also a bad option.
>> >
>> > Is there any reason not to make the schedule
>> > starting with CA?
>> 
>> I believe the reason was that we are in CA in December. The
>> problem with making the swap is that we then end up likely
>> to do it again since we are in CA in December by the schedule
>> below. We then end up in California 50% more often...
>> 
>> > Feb 07-09 California
>> > Mar 28-30 Chicago
>> > May 09-10 California
>> > (not May 16-17 due to IPDPS)
>> > (not May 23-25 due to CUG)
>> > Jul 18-20 Chicago
>> > Sep xx-yy together with EuroMPI
>> > (not Nov 13-18 ??? due to SC - who knows exact date?)
>> 
>> The exact dates are Saturday November 12 - Friday November 18.
>> Since the first day is just the education program, your dates
>> are probably close to accurate for most.
>> 
>> Bronis
>> 
>> 
>> 
>> 
>> > Dec 05-07 California
>> >
>> > I hope, I have not moved some dates without need,
>> > i.e., correctly representing the state of current
>> > discussion.
>> >
>> > Best regards
>> > Rolf
>> >
>> > ----- Original Message -----
>> >> From: "Jeff Hammond" <jeff.science at gmail.com>
>> >> To: "Main MPI Forum mailing list" <mpi-forum at lists.mpi-forum.org>
>> >> Sent: Thursday, July 8, 2010 4:24:36 PM
>> >> Subject: Re: [Mpi-forum] 2011 meetings
>> >> I find Chicago quite pleasant except for December 15 to January 5,
>> >> when I know the Forum will not meet. While it may be cold, there is
>> >> frequently plenty of sunshine. If you're inside a building, it
>> >> might
>> >> as well be Hawaii outside.
>> >>
>> >> I plan to never attend a meeting in Atlanta because my past
>> >> experiences with Delta and/or that airport have all been terrible.
>> >>
>> >> Jeff
>> >>
>> >> On Thu, Jul 8, 2010 at 9:11 AM, Jeff Squyres <jsquyres at cisco.com>
>> >> wrote:
>> >> > On Jul 8, 2010, at 10:01 AM, Torsten Hoefler wrote:
>> >> >
>> >> >> > At this stage I will keep that meeting in Atlanta, unless more
>> >> >> > people
>> >> >> > speak up.
>> >> >> I think it is unreasonable to expect that a majority of the
>> >> >> Forum
>> >> >> will
>> >> >> speak up via email (or even read this thread :-). It would be
>> >> >> reasonable
>> >> >> to do statistical sampling. I heard 4 people against Atlanta and
>> >> >> 0
>> >> >> in
>> >> >> favor so far. Speak up!
>> >> >
>> >> > I prefer California and Chicago to Atlanta.  Atlanta is certainly
>> >> > not a deal-breaker for me, but I do prefer the other two.
>> >> >
>> >> > --
>> >> > 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
>> >> >
>> >>
>> >>
>> >>
>> >> --
>> >> Jeff Hammond
>> >> Argonne Leadership Computing Facility
>> >> jhammond at mcs.anl.gov / (630) 252-5381
>> >> http://**www.**linkedin.com/in/jeffhammond
>> >>
>> >> _______________________________________________
>> >> 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)
>> >
>> > --
>> > 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
>> _______________________________________________
>> 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


More information about the mpi-forum mailing list