Software Coordinating Committee Conference Call
		       December 6, 2001 4:30 PM EST
			  Recorder: C. DeTar

Present:

  Brower, DeTar, Pochinski, Edwards, Gottlieb, Mendez, Simone, 
  Mawhinney

Absent:
 
  Holmgren
  Watson

     1. Review (quickly) the status of the goals set at end of
        the Jlab workshop

      --Level 1 QMP-API status

        Edwards: New version of MP API is now on web site.
           MPI implementation is running.  Under development
           is a gm implementation at essentially zero copy
           but not global reductions.
           Will send URL.

      --MILC/Level 1 port

        DeTar: On hold waiting for new version.

      --QCDOC implementation of Level 1

        Mawhinney: Chulwoo working on lower layer (DMA), so not up
           to Level 1, yet.  Limited by development of hardware.
           No QCDOC-required functional changes in QMP API 
           anticipated at this point.  QCDOC implementation
           by 1st weeks of January.

      --MILC on QCDOC simulator
 
        Mawhinney: Upgrading simulator software from v 2.1 -> 2.3 plus
           enhancements by Boyle and Joo.  Now stable and Calin
           will get together with Doug Toussaint to resume
           testing MILC code.

     2. Discussion of Level 2 Data Layout Standards

       -DeTar: Described proposal circulated
       -Brower: Set of standard types:  Wilson fermions
        lattice-wide, red, black - four components or two.
       -DeTar: Is this flexibility excessive, or should we
        be less ambitious?
       -Edwards: Proposal is putting too much into level 2.  Pick any
        format, but pick one.  Do conversion for Dirac operators.
       -Mawhinney: Keep Level 2 basic and efficient.  
       -Brower: Worry about complexity of level 2 and allowing
        user to hang himself too easily.
       -Brower: Level 2 code is for efficient coding of noncritical
        parts of the code.  Can we implement a fixed layout
        but leave in the flexibility?
       -Brower, Edwards, DeTar to continue discussion.
       -Mawhinney: How to reach closure?
**      All to think more about this during the coming week.
        
Next call Thursday, December 13, 3 PM EXT

Conference concluded at 6:00 PM EST.