LHC Computing Grid Project

Project Execution Board

Notes of the meeting of Tuesday January 27, 2004

DRAFT 2  8/FEB/2004

 

Present:

Dario Barberis, Federico Carminati, Philippe Charpentier, David Foster, Bob Jones, Jürgen Knobloch (secretary), Alberto Masoni, Mirco Mazzucato, Bernd Panzer, Les Robertson (chair), Zdenek Sekera, David Stickland

Phone: Nick Brook (from 17:10)

 

 

Actions: Actions are identified by bold blue italics.

Minutes of last meeting and matters arising

The minutes of the meeting of the 12th January were accepted.

Matters arising:

Gag mandate: Federico has made slides available (transparencies). The membership is being completed in a mail exchange between Federico and the experiments. The proposal was approved.

Major decisions from recent meetings

Architects Forum - Torre

Licenses of commercial software will expire soon (HEP-wide NAG-C in June; Iris Explorer/ Open Inventor CERN site license in December 2004). Andreas Pfeiffer is preparing a proposal on what to do.

A discussion took place on the ATLAS and LHCb positions concerning a POOL relational back end in view of conditions database implementations. Dirk Duellmann continues to talk to the experiments in preparation of an updated POOL workplan.

LHCb has almost achieved the POOL persistency milestone (waiting for Windows binaries). Philippe confirmed that this is also OK now - since January 26.

A draft SEAL workplan was presented at the Architects Forum by Pere. Vincenzo wanted to see agreement by the experiments item by item. An important item was the question of a common dictionary between ROOT and LCG. Vincenzo did not want to see one and only one dictionary implementation (based on CINT).

Waiting for the POOL draft plan for 2004 – feedback from the experiments is required.

Grid Deployment Board – Mirco

Mirco summarized the items that will require some follow-up:

Dave Kelsey has presented the status of user registration/authorization in the context of virtual organizations (VOs).

Les announced that for CERN Ian Bird has been given the job of discussing with the experiments and the administration to prepare a single system for authorization in the experiments and at CERN in general. This would not prejudice the responsibility of implementation.

John Gordon has presented the status and plans for the Grid Operations Centre (GOC). An action item is: [text to be provided by Mirco]

The question of interfacing the grid user support with experiments support will be discussed in a meeting between Karlsruhe and representatives from the experiments.

Another action item is to set up a group working on the interoperability between LCG and Grid-3.

The proposed schedule for data challenges in 2004 was endorsed.

Bernd should now request the information on computing resources at centres in order to complete the centrally maintained tables. 

Summary, conclusions and follow-up from ARDA workshop

Les sent out note (document) giving his view of the outcome of the workshop. The discussion circled around the questions what the scope of the ARDA project was, whether a project leader was required, and what the relation to EGEE was.

After a lengthy discussion, there seemed to be consensus on the following points:

  • What is expected by the experiments is an end-to-end solution based on LCG, VDT, and Alien. Four instances using as much as possible common software should be available in prototype form within six months – therefore it is essential to start now.
  • The basic middleware will be produced in the structured EGEE activity under the leadership of Frédéric Hemmer.
  • The middleware experts having a large fraction (60-80%) of their time available should work in the middleware project. The assignment should be based on competence.
  • It is essential that people from the experiments participate actively in building the prototypes based on this middleware and products from the applications area such as POOL. There will be 4 people hired in EGEE to be complemented by 4 people from the PH department.
  • ARDA is seen as a jumpstart for EGEE.
  • The leader of the ARDA project should have a mostly coordinating role making sure that things work together and finding out where difficulties are.

 

A number of people from the experiments announced possible contributors to the project. To clarify the meaning of the term ARDA, Les proposed to call the middleware ARDA.

Les will write a revised version of his note taking this discussion into account.

 

Collaborative tools - status and conclusion on RTAG

Dario has updated the proposed mandate (transparencies) of the RTAG and called for comments. As there were no further comments and the PEB agreed the proposal was adopted. Also the proposal to nominate Steve Goldfarb from University of Michigan as chairperson was endorsed.

Interested parties such as CERN/IT, EGEE, and VRVS will nominate members. The experiments are already advanced in nominating their representatives:

ATLAS: Roger Jones and Steven Goldfarb

CMS:

LHCb: Ian McArthur (Oxford) and Gerhard Raven (NIKHEF)

Alice:

It is expected that a report will be delivered within two months.

LCG-2

David Stickland reported a technical problem that POOL and the replica manager use file catalogue differently.

In addition, the replica manager can currently only handle lower casing while CMS have mixed cases. These problems represent show-stoppers for CMS to use LCG-2. Dirk and Ian are investigating the problem and will come back with solutions to CMS.

It is proposed to handle problems like these in a regular (weekly) “Deployment Meeting”.

AOB

 

  • Bob Jones announced that an EGEE all-activity meeting took place 13 & 14th January. A follow-up meeting is tentatively scheduled for March. He also informed that in a letter from Brussels signature of the EGEE agreement was envisaged before end January.
  • Federico announced that the team from Trento that has developed the code checking tool is proposing a follow-up project. He will send the proposal to the PEB. We will then decide whether we go along and if necessary how to find the required funding.

 

 

Actions

#

Date opened

Description

Responsible

Date closed

1

16dec03

ALICE, CMS and LHCb to name someone responsible for coordinating deployment on LCG-2

Federico, David S., Philippe

 

2

16dec03

Understand why the substantial resources in Liverpool are not available for LCG-2.

6jan04- visit to RAL organised for 24jan04

Les

 

3

16dec03

Confirm that the absence of BNL in the LCG-2 deployment list is due to manpower shortage

Les

 

4

16dec03

Experiments to request through their national contacts that their resources in the core LCG-2 centres are integrated in LCG-2

Federico, Dario, David S., Philippe

 

5

16dec03

Regional centres to be asked to clarify their mass storage plans.

Presented by RCs in GDB of 13jan04

Les

13jan04

6

12jan04

Revised proposed GAG mandate

Federico