[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [mpi-21] SC'07 MPI Forum organization meeting
I'm from Europe and I'm not against your proposal.
I only want to describe my positive experience with travelling every
6 weeks to the conference Hotel at Chicago O'Hara Intl. Airport:
There should be always a flight from all over US and Europa
to arrive in Chicago O'Hara before 1:30PM.
MPI-2-meeting starting at 1:00PM was a little bit early for me,
(but okay) 1:30PM or 2:00PM would have been better.
Currently, there are 6 flights from Europe arriving between
11:25AM and 12:35PM (from Munich, Frankfurt Düsseldorf,
Amsterdam, Stockholm, Dublin),
another 6 flights from 1:00PM to 1:20PM (from 3x Manchester, London,
Brussel, Amsterdam),
and 5 flighte arriving 2:00PM - 2:20PM (Dublin, Vienna, 2x London,
Madrid)
>From airport to the hotel was a 5 min trip with the Hotel shuttle.
And after the meeting (ending at noon), two or three hours later
is your flight back direct to Europe with one additional hop in Europe
and at about 9AM you're back with a short night on the flight.
An important issue: The schedule for several next meetings was always
early available. Therefore I could use always two cheap crossed
tickets for two consecutive meetings.
I expect, that to be always at the same meeting place, minimizes all
organizational things.
Best regards
Rolf
On Thu, 18 Oct 2007 11:45:10 +0100
"Supalov, Alexander" <alexander.supalov@xxxxxxxxx> wrote:
>Hi,
>
>I think this is a good idea if we agree to hold a couple of meetings in
>Europe and Asia, to make the process equally "painful" to everyone.
>
>If that's not an option, I think we should consider relaxing the rules a
>little to allow remote attendance, apart from the kickoff meeting and
>may be the some milestone meetings, like the final one.
>
>Looking for support from those who will suffer most. ;)
>
>Best regards.
>
>Alexander
>
>--
>Dr Alexander Supalov
>Intel GmbH
>Hermuelheimer Strasse 8a
>50321 Bruehl, Germany
>Phone: +49 2232 209034
>Mobile: +49 173 511 8735
>Fax: +49 2232 209029
>
>
>-----Original Message-----
>From: owner-mpi-21@xxxxxxxxxxxxx [mailto:owner-mpi-21@xxxxxxxxxxxxx] On
>Behalf Of Cownie, James H
>Sent: Wednesday, October 17, 2007 11:48 AM
>To: mpi-21@xxxxxxxxxxxxx
>Subject: RE: [mpi-21] SC'07 MPI Forum organization meeting
>
>> * Form of effort - all electronic, mix of electronic and periodic
>> face-to-face meetings, or some other manner.
>> * Voting procedures - who can vote, how are votes conducted, etc.
>> * Organizational structure - Identify those who will handle forum
>> logistics.
>
>Having done HPF, MPI-1 and MPI-2, and seen how OpenMP works, I would
>strongly recommend sticking with the original MPI organizational style.
>
>So
> 1) Have a relatively short timescale (a year at most), so you should
>be
> aiming to have the work completed to present at SC08.
> 2) Have regular face-to-face meetings (MPI-1 was every 6 weeks, MPI-2
> similar, IIRC).
> 3) Have voting as one vote per qualifying organization. Qualifying
>means
> "the organization was represented at 2 of the last 3 meetings
> (including this one)". An organization is defined sensibly, so each
>
> national lab gets one vote (they're not all DoE), and each company
>or
> university gets one vote.
> 4) Attendance at the meeting is the only qualification criterion for
> voting. (No membership dues, beyond the meeting fees which are just
>
> to cover meeting rooms, coffee, hard-copies of materials etc).
>Meetings
> are open to anyone who is prepared to come.
> 5) Material has to be approved twice (at two separate meetings) to get
>
> into the standard.
>
>The aim is to ensure that people are committed to working on this
>seriously. If they're giving up one week in six that requires much more
>effort than an hour's teleconference once a week. The grueling schedule
>is enough to keep attendance to a reasonable level. You want to
>encourage organizations to send the same people to each meeting, so that
>you get a group which gels as the MPI forum, who know and respect each
>other.
>
>Of course MPI-1 and MPI-2 were before the advent of wi-fi. You might
>want to find meeting rooms which specifically do NOT have wireless
>access :-)
>
>Basically this isn't intended to be a pleasant experience. By the end of
>it you want everyone to be glad it's over :-)
>
>-- Jim
>
>James Cownie <james.h.cownie@xxxxxxxxx>
>SSG/DPD/PAT
>Tel: +44 117 9071438
>
>
>
>
>> -----Original Message-----
>> From: owner-mpi-21@xxxxxxxxxxxxx [mailto:owner-mpi-21@xxxxxxxxxxxxx]
>On
>> Behalf Of Richard Graham
>> Sent: 12 October 2007 17:48
>> To: mpi-21@xxxxxxxxxxxxx
>> Subject: [mpi-21] SC'07 MPI Forum organization meeting
>>
>> As previously announced, we will be holding a meeting on Wed. Nov.
>14th,
>> from 5:30 pm to 7:00 pm in room D6 at the Sparks convention center in
>Reno
>> -
>> site of SC'07. The intent of the meeting is to organize for an active
>> period of the MPI forum. The proposed agenda is:
>>
>> * Introduction: why restart the forum discussions
>> * Scope of effort - discussion. There is a wide range of opinions on
>how
>> to proceed, with a clear need for clarifications and simple
>corrections to
>> the 2.0 standard (such as changing the size of the length parameter
>used
>> in
>> MPI send functions), as well as a desire for larger changes to the
>> standard,
>> such as adding functionality to support process fault tolerance. We
>need
>> to
>> decide if to consider both types of of changes in a single standard
>> change,
>> as parallel but distinct set of changes to the standard, or some other
>> approach.
>> * Form of effort - all electronic, mix of electronic and periodic
>> face-to-face meetings, or some other manner.
>> * Voting procedures - who can vote, how are votes conducted, etc.
>> * Organizational structure - Identify those who will handle forum
>> logistics.
>>
>> If time permits:
>> * List of initial topics of interest. I will compile a list of
>initial
>> items, so please send topics of interest to mpi-21@xxxxxxxxxxxxx .
>>
>> Please feel free to comment on the agenda, and more important, please
>plan
>> to attend.
>>
>> Thanks,
>> Rich Graham
>---------------------------------------------------------------------
>Intel Corporation (UK) Limited
>Registered No. 1134945 (England)
>Registered Office: Pipers Way, Swindon SN3 1RJ
>VAT No: 860 2173 47
>
>This e-mail and any attachments may contain confidential material for
>the sole use of the intended recipient(s). Any review or distribution
>by others is strictly prohibited. If you are not the intended
>recipient, please contact the sender and delete all copies.
>---------------------------------------------------------------------
>Intel GmbH
>Dornacher Strasse 1
>85622 Feldkirchen/Muenchen Germany
>Sitz der Gesellschaft: Feldkirchen bei Muenchen
>Geschaeftsfuehrer: Douglas Lusk, Peter Gleissner, Hannes Schwaderer
>Registergericht: Muenchen HRB 47456 Ust.-IdNr.
>VAT Registration No.: DE129385895
>Citibank Frankfurt (BLZ 502 109 00) 600119052
>
>This e-mail and any attachments may contain confidential material for
>the sole use of the intended recipient(s). Any review or distribution
>by others is strictly prohibited. If you are not the intended
>recipient, please contact the sender and delete all copies.
>
Dr. Rolf Rabenseifner . . . . . . . . . .. email rabenseifner@xxxxxxx
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)