Hi  everybody,

Thank you for your time today. It was a very good discussion. Here's what I captured (please add/modify what I may have missed):
 
Present: Leonid Meyerguz (Microsoft), Rich Graham (ORNL), Richard Barrett (ORNL), Torsten Hoefler (ISU), Alexander Supalov (Intel)
 
- Opens & introductions 
 
- Scope of the effort 
  - Rich
    - Minimum subset consistent with the rest of MPI, for performance/memory footprint optimization
    - Danger of splitting MPI, hence against optional features in the standard
    - Both blocking & nonblocking belong to the core
  - Torsten
    - Some collectives may go into selectable subsets
    - MPI_ANY_SOURCE considered harmful
  - Leonid
    - Flexible support for optional features, means for choosing and advertising level of compliance/set of features
  - See enclosed email for Alexander's POV
 
- General discussion snapshots
  - Support of subsets: some or all? If some, possible linkage problems in static apps (or dead calls). If all, where's the gain?
  - Optional: really optional (may be not present) or selectable (are present but may be unused)?
  - Performance penalty for unused subsets: implementation matter or standard choice?
  - Portability may be limited to certain class of applications (think FT, master-slave runs)
  - All we design needs to be implementable, complexity needs to be controlled
  - An ability to use certain set of subsets should not preclude pulling in other modules if necessary
  - Whatever we do, it should not conflict with the ABI efforts
  - Need to stay nice and be nicer wrt to the libraries (think threading) and keep things simple
  - The simplification argument, if put first, may not be liked by some
 
- Next steps
  - Please comment on these minutes, and add/modify what I may have missed
  - I'll prepare a couple of slides by next week summarizing our discussion so far; again, your feedback will be most welcome
  - At the meeting, it may be great to meet F2F briefly and discuss any eventual loose ends before the presentation at the Forum; I'll see to this
 
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
---------------------------------------------------------------------
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.