US ATLAS Computing meeting
March 20, 2002

Reported by David Adams
Updated 20mar02 1350

Attending: David Adams (BNL), Alex Undrus (BNL)
Rob Gardner(IU), John Huth (Harvard) Ian Hinchliffe (LBNL),
Torre Wenaus (BNL), Jim Shank (BU)
and many more

************************************************************************
Agenda
************************************************************************
    - Status of the project: funding and its implications
      If the managers (John, Jim, Howard) attend; I have asked them to
    - Other management news
    - Facility news, issues, user feedback
      - Tier 1
      - Tier 2
    - Software support news, issues, feedback
    - Data Challenge status and issues
    - Reactions, discussion on LCG launch week
    - Core software activity summaries
      - Universities
      - ANL
      - BNL
      - LBNL
************************************************************************
Funding
************************************************************************

Huth

There is a DOE funding shortfall for the second half of this year--
expected 1.25m$, will get 1.0 M$. In addition NSF shorfalls leave us
0.6-0.7 M$ short. He expects to get some additional NSF funds and to be
left with losing two FTE's.

He will circulate an email before he leaves (for a week) on Friday.

************************************************************************

Torre

We should add sustained support for training in our budget,
specifically the web training work at Michigan.

************************************************************************
Facility news
************************************************************************

No Tier 1 represented.

************************************************************************

Rob Gardner - IU

Doing PACman installation of VDT 1.0. Recent email encouraged people
to try it out.

************************************************************************

Jim Shank - BU

Working on PACman. Working with Christian to get dependency information
from CMT. Also with Pavel to get is scripts.

They are installing hardware but he did not describe it.

Torre - We should install a real application.
Jim - We can run ATLFAST.

John - May be delays for DC1-2
Ian - DC1-1 will be delayed

************************************************************************
Software support
************************************************************************

Alex Undrus - BNL

3.0.0 installed at BNL.

Implementing testing for nightly builds including component testing and
athena testing. A similar effort is taking place at CERN by
Jean-Francois Laporte.

************************************************************************
Data challenges
************************************************************************

Ian - LBNL

Still working on April 15 timescale. Only one tag before then. Core
packages need to be in place next Wednesday. Every one else has 10
days.

David Q. - We have many things to put in. Plowing ahead anyways.
David M - Are these thing required for trigger production.
David Q. - No.

************************************************************************

David Malon - ANL

For DC1-1 want to support ROOT IO: generator and RD event. Converters
in place; BNL is working to read the generator output into ATLSIM.
Aggresive time scale; we will probably make it.

Ian - Looking for Tier 2 sites to work on event generation for DC1-2.
      Needs people, not just machines.
Ian - Sites should generate events of the type of physics that interest
      them.

************************************************************************
LCG workshop
************************************************************************

See workshop summaries on the last day.

John - What is plausible or implausible to do in concert with other
       experiments?
Torre - CMS will not use Gaudi/Athena in the short term.
Torre - CERN will provide support for Rene Brun for ROOT.

************************************************************************
Core sofware activities.
************************************************************************

Torre - There will be a US-ATLAS GRID testbed meeting at UTA April 4-5.

David Q - Detector description meeing April 9-10. Not public.
David Q - Andrea Dell Acqua will be present.
Torre - What's on the table?
David Q. - Thinking about API. Probably not scrap XML. What is needed
           for DC1-2?

************************************************************************

David M (ANL)

There will be a common project in data management. He is optimistic
about delivering a specification for persistency. A streaming layer
consistent with root with a lightweight cataloging.

Making NaiveOby examples generic.

Torre - Should make connection with the HES work at BNL.
Ed Frank - Still considering going to BNL.
         - three pieces: ADB document, GRID and HES
         - Want to expand ADB to include some GRID concepts.
         - Which new ideas in HES should move up into the ADB.

************************************************************************

David A (BNL)

Waiting for comments on HES. Development proceeding to better
understand the interface.

************************************************************************

David Q. (LBNL)

Much activity at CERN on the EDM.

Testing Gaudi 9.0.1.

Much work going on with testing.

Andrea Dell Acqua is fully supportive of integrating GEANT 4 into
athena.

************************************************************************