LHC Computing Grid Project
Project Execution Board
Notes of the meeting of Tuesday
November 4, 2003 - Draft
Present:
Minutes of previous meeting(s) (October 14) and matters arising
The minutes of the last meeting incorporating a change proposed by Ruth
were approved. The new version is available here.
Status and Update on High Energy Physics Joint Technical Board and Grid
2003 (Ruth Pordes) ( 
transparencies )
The Intergrid Joint Technical
Board (JTB) was established 2 years ago. The first year goals were defined by “Glue”
and aimed at enabling interoperability between the EU and
As next steps, the membership of the JTB will be extended, the
work on items from Glue Phase II which has started will be continued, and a workshop
will be held in March 2004 in
Grid2003 (Grid3) is a coordinated project between iVDGL, GriPhyN, PPDG, and the physics experiments, principally being led by USCMS and USATLAS. The Grid3 environment is now deployed on 25 sites with about 1500 processors. Targeted tests with LCG for ATLAS and CMS are starting. At Fermilab, local discussions address the question of common grid interfaces across groups such as Run II using SAM and CMS using VDT.
In the discussion, the question of “namespaces” (experiment-wide
vs subgroups) was raised. The JTB is considered a
good place to discuss technical matters and propose solutions which should then
be forwarded to the respective communities. It is recognized that there are two
strong communities – OSG and EGEE. A way on how to pull together SAM and RLS is
not obvious.
Proposed
reorganisation of PEB and SC2 ( document )
Les
presented a proposal to re-organize the PEB and the SC2. The proposal comes in
two flavors depending on whether SC2 continues to organize reviews or the
review process is taken over by the LHCC. The problem with the current PEB is
that the coordinators are not present and therefore decisions on certain things
cannot be taken at the meeting. In the PEB - renamed Project Management Board
(PMB) - the computing coordinators and the area managers would be present.
The general
feeling was that it is a good thing to improve the decision-making process by
including the coordinators.
The
discussion circled around the question of representation of regional centres. It was agreed that having all regional centres represented individually would make the meeting too
big and that a “smallish” number of people was ok. The proposal of having two
or three people representing the regional centres was
felt appropriate though identifying the right people will be difficult. One
possibility would also be that just the GDB chairperson would represent the centres and regions and act as the link between the PMB and
GDB.
The
discussion will continue in the SC2 and a decision will be taken at the next
POB.
AOB
Actions from Computing RRB
Chris
reported that a task force to create a draft of the computing MoU for LCG will start now. Initially, until end 2003, the
task force will be led by Hans Hoffmann. The LCG MoU
will be one of five – the others covering each of the experiments. It will be
important to watch that the LCG MoU is compatible
with experiments MoUs. As next step Hans Hoffmann will
send the mandate to CRRB and ask for nomination of members of the task force. He
will then select the membership for final agreement.
It is expected
that the task force will produce a preliminary text by the time of the CRRB on
In the RRB
there is increasing pressure to replace the CERN-centric presentation by
presentations from the other centres.
The question
was raised whether the manpower report should be limited to CERN or whether the
outside situation should be also reported.
Internal project review
The LCG internal
review for Grid and Fabric is now organized for October 17-19 and the session outline
is available from the PEB
agenda pages. The experiments
will be asked short presentations on how they plan to arrive at the computing
models.
LHCC comprehensive review
The schedule
is also available on the PEB agenda pages.