Updated protocol. Tickets 185, 207, 215, 254, 255, 262, 263, 267, 268 and 312 on Slide 9 are now all prepared for reading. Tickets 162, 182, 219 on Slide 11 also. Best reagrds Rolf ----- Original Message -----

From: "Rolf Rabenseifner" <[email protected]> To: "Main MPI Forum mailing list" <[email protected]> Sent: Monday, January 16, 2012 4:46:05 PM Subject: [Mpi-forum] Protocol of MPI-3 ticket action plan for March 2012 - Both parts Dear all ticket owners and chapter authors,

Here is now the complete protocol combined with all updates for the ppt as far as I know about. The ppt and pdf contain our route 66 (tickets) to MPI-3.0. The txt contains the MPI-next and withdrawn.

All orange tickets with "!" still need some work.

I expect that this will be the complete working plan for the March meeting. For the agenda, which will have a maximum of 8+12+6=26 hours, Rich will need your time-estimates.

My statistics and estimates: - 12 dark green tickets = passed = 0 h - 2 light green tickets (299,229) = 2nd vote needed = 10 min - 8 blue tickets (272,168,298,308,309,284,280,294) = 1st vote needed = 45 min - 11 orange simple tickets on slide 9 = reading+ONE VOTE = 35 min Sum = 1.5 h

- Slide 2, ticket 273 = partial re-reading = 0.5 h - Slide 3, small tickets 275+283+300 = reading = 0.5 h - Slide 4, FT (see at the end of the list) - Slide 5, ticket 271 = reading = 0.7 h - Slide 6, tickets 310+311+217 = partial re-readings = 2 h - Slide 7, tickets 286+305 = (partial) re-reading = 0.5 h (first day) discussion probably off-line, re-reading = 0.5 h (last day) - Slide 7, ticket 264 = partial re-reading = 0.5 h - Slide 8, ticket 278 = partial re-reading = 0.5 h - Slide 8, ticket 281 = full re-reading = 0.5 h - Slide 8, ticket 303 = discussion ? = 1 h - Slide 9, already included in the first part of the list - Slide 10, readings: ticket 158 (10min), 187 (30min), ticket 227 (10min), 256 (10min), 259 (40min), + unexpected discussions (20min) = 2 h - Slide 11, reading+ONE VOTE: tickets 162, 182, 219 = 22 small items = 0.5 h - Slide 11, reading+ONE VOTE: tickets 125+126 = many items but all verified by reviewers = 0.5 h - Slide 11, reading+ONE VOTE: ticket 218 (may include discussion) = 1 h - Slide 12, reading of ticket 159 + discussion : 1 h - Slide 12, reading of ticket 163 + discussion : 1 h - Slide 12, reading of ticket 194+195 + discussion : 2 h - Slide 12, reading of ticket 222 + discussion : 1 h - Slide 12, reading of ticket 293 : 1 h - Slide 12, reading of ticket 304 + discussion : 1 h Sum = 32 orange tickets = 18.2 h

- Remaining time for FT: 6.3 h

I hope that my estimates fit to the estimates of the ticket owners. And that we spare some additional time in some of the tickets to get some additional breaks, especially on Tuesday. For Tuesday, I recommend to add at least three breaks 15 / 30 / 30 min to the agenda and to lengthen it to 10pm to keep the working time. (On Monday, I personally would really like to stop at 9pm due to oversee timezone switch).

We should try to get really all through, because the group in Japan will be too small for significant discussions.

Goal at end of March-meeting: - 12 tickets passed with 2 votes - 8 tickets passed with 1st vote (2nd vote needed in Japan) - 17 text-only or trivial tickets passed with only one vote - 26 tickets passed formal reading and ready for 1st vote in Japan - FT with about 3 tickets, passed formal reading Goals for FT: a first stage that help the next years and may need upgrades in some years (as we have seen with 1-sided in the two steps in MPI-2.0 and 3.0). -------- This is the total of our route 66 (tickets) for MPI-3.0

Best regards Rolf

----- Original Message -----

...

From: "Rolf Rabenseifner" <[email protected]> To: "Main MPI Forum mailing list" <[email protected]> Sent: Thursday, January 12, 2012 2:12:42 AM Subject: [Mpi-forum] Protocol of MPI-3 ticket action plan - Part A Hi all,

here is the text file to you all - this is Part A. Part B is coming later. This will be the commented ppt.

Best regards Rolf

-- Dr. Rolf Rabenseifner . . . . . . . . . .. email [email protected] 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 [email protected] http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi-forum

-- Dr. Rolf Rabenseifner . . . . . . . . . .. email [email protected] 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 [email protected] http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi-forum

-- Dr. Rolf Rabenseifner . . . . . . . . . .. email [email protected] 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)