Dear All,
Please find attached the weekly GridPP Project Management
Board Meeting minutes. The latest minutes can be found each week in:
http://www.gridpp.ac.uk/php/pmb/minutes.php?latest
as well as being listed with other minutes at:
http://www.gridpp.ac.uk/php/pmb/minutes.php
Merry Christmas and a Happy New Year, Tony
________________________________________________________________________
Tony Doyle, GridPP Project Leader Telephone: +44-141-330 5899
Rm 478, Kelvin Building Telefax: +44-141-330 5881
Dept of Physics and Astronomy EMail: [log in to unmask]
University of Glasgow Web: http://ppewww.ph.gla.ac.uk/~doyle
G12 8QQ, UK Video - IP: 194.36.1.33
________________________________________________________________________
GridPP PMB Minutes 198 - 19th December 2005
============================================
Present: John Gordon, Sarah Pearce, Tony Doyle, Dave Britton, Robin Middleton,
Dave Newbold, Steve Burke, Tony Cass, Steve Lloyd, Roger Jones, Peter Clarke
Apologies: Dave Kelsey, Jeremy Coles
1. Quarterly Reports [DB]
=========================
DB said that Q3 was almost finished but RM said he was still working on
finalising the Security report with Andrew McNab at the moment. TD wished
to note that this was the worst period of quarterly reporting (in terms
of timeliness of reporting and feedback) in GridPP history. He hoped
it would improve with 05Q4 (needed as input to the next OC).
The PMB pages at
http://www.gridpp.ac.uk/pmb/
now include the following info.
"Quarterly reporting is used throughout the project. At the end of each
quarter the following deadlines apply:
# within 2 weeks - report from Work Package Leader
# within 3 weeks - feedback from the Area Manager (where required)
# within 4 weeks - feedback from the Project Manager (where required)
PMB review takes place at the next available Monday meeting"
2. F2F agendas [DB, DK]
=======================
DB will make a start in compiling the agenda as he has been unable to
arrange discussion on this with DK.
3. CHEP06 status [RM]
=====================
RM had emailed everyone concerned and had received some visit notices.
The budget allocation is 22k. He said he hoped to receive 50% funding
from EGEE for some of the people from RAL who are attending CHEP.
4. User Board Review [DN]
=========================
DN discussed the following report which he had emailed previously:
The response to the UB requirements poll was near-complete. DN sought
a great deal more information than in previous rounds, predicting the
need to make 'hard decisions' this time around, and this bore fruit,
with several experiments providing a great deal of information on their
plans and requirements for the coming year.
- The meeting was well-attended, though unfortunately no BaBar
attendance. Particularly pleasing to see good attendance by the smaller
experiments (MINOS, SNO, UKQCD).
- The UB viewed its role as being to make allocations of CPU, disk,
tape at Tier-1, and outline allocations of CPU, disk at Tier-2. No
attempt was made to understand Tier-2 usage by site.
- All resources were undersubscribed, or only mildly oversubscribed,
with the exception of Tier-1 disk. This is with respect to the
'availability' numbers supplied by Andrew Sansum (for Tier-1) and Steve
Lloyd (for Tier-2). We noted that the availability numbers have now
understandably moved away from the v26b planning. We have some trouble
discerning the level of reality of the Tier-2 numbers, compared to what
is available *via LCG*.
- It's good to see that the experiments have taken the call to request
Tier-2 resources seriously. However, there is a clear perception
(stated at the meeting) that Tier-2 disk is not viewed as reliable.
- The Tier-1 disk is very heavily oversubscribed - total requests for
Q1 were 287.5TB, against an allocatable total of 179TB. The largest
requests were BaBar (120TB) followed by Atlas, CMS, LHCb (40, 40, 10TB)
who require disk to support data challenges early in 2006. On current
planning, the problems gets worse in Q2. In DN's view, it is neither
practical nor sensible to overallocate disk.
- DN applied the following algorithm: the 'small' experiments, including
ALICE, get to first order what they asked for. We reappraised with CDF
the reality of their 30TB request, and this was reduced to 10TB. Then
the question is one of balance between BaBar and LHC. We recognised
that the planning v26b gives a clear message on BaBar disk allocation;
however, we essentially decided to pass this up for a 'strategic'
decision, in the first instance to the PMB, and hence to the Tier-1/A
Board on January 16th.
- Following today's discussion, we must make public allocations to
enable progress in the early New Year. It was noted that these would
need to be endorsed at the Tier-1/A and Tier-2 Board meetings.
- The other main topic to bring to PMB is the question of experiment
responses to the questionnaire, and actions to be taken. Without a
face-to-face meeting, no major progress was made on this. It was agreed
that the UB chair and deputy chair would begin the process of
discussing with the main respondents to the questionnaire whether any
tangible progress had been made since last year in their areas of
concern. The progress should be documented.
- One immediate response was that the portal does not seem have had the
take-up which might have been hoped, which certainly has the potential
to affect our ambition for Grid-only Tier-1 use, and more extensive
Tier-2 use. DN needs to compare the plans in this area with the
milestones.
DB noted shortly after the meeting the various interactions by Gidon Moont
with the other experiments
MICE, CALICE, UKDMC, T2K, PHENO, DESY (site rather than specific
experiments), GridCC and (possibly) MINOS.
The message to the UB is that:
a) There is a portal toolkit available.
b) If you contact Gidon he will gladly give you help getting started.
c) .... but he cannot undertake to develop an experiment-specific portal
for you at present.
d) There was a demo and presentation at the Durham GridPP meeting which
can be consulted.
e) There will be further information at GridPP15.
TD noted that there was a meeting with BaBar at 5:15pm today to discuss
Tier-A problems and he will send a summary of the outcome to the PMB.
5. GridPP15 status [JG]
=======================
JG said that 55 people had registered and he would email everyone to
remind them that the deadline for registration is today. Those attending
will be assigned hotels this week. Sponsorship for the dinner has not
been arranged. The PMB meeting will be held in the same room as the main
conference.
6. AOCB
=======
TD said that the PMB's admin. assistant, Barbara Scott, was leaving to
take up another position at Glasgow University in January. The PMB thanked
Barbara for three years' support for the project and wished her well in
her new job.
The next face-to-face PMB meeting will be on 10th January at RAL prior to
GridPP15.
TD decided to pass on the Standing Items and Actions due to time
constraints today at the last meeting of the year.
STANDING ITEMS
==============
SI-1 ALL: to provide any news items and meeting dates for the web page to SP.
SI-2 Production Manager's weekly report of issues.
JC was unable to connect to the meeting. The main area of current interest
relates to the transfer tests. A summary of the situation can be found
here:
https://wiki.gridpp.ac.uk:443/wiki/Service_Challenge_Transfer_Tests
Other areas of note are:
1) Security challenges - almost complete now and have proved
valuable in revealing problems with our tracking processes as well as
at some sites tracing problems.
2) Use of Tier-2 resources (in particular storage) - this relates
to issues raised at the last UB (DN has the details).
3) Enabling new VOs - we already knew that the process has to be
made smoother but will ensure the process is clearer soon after a
number of new requests.
4) The vulnerability follow up is still a concern. Site Admins have a
genuine local concern and may not follow a requested resolution route if
they feel nothing is happening. This will need to be taken up at the T2B
in January.
SI-3 LCG Management Board Report of Issues.
SI-4 Documentation Officer's weekly report of issues.
ACTIONS AS AT 19TH DECEMBER 2005
===============================
184.1: TC to write document "How CERN helps the small sites to install
and manage the LCG software".
- ongoing
184.3: RM and RJ to document Gap Analysis
- ongoing
184.4: DN to document UB questionnaire issues
- ongoing
187.1: JG to prepare combined actions list from GridPP14 meeting
- ongoing
197.1: SL to determine realistic estimate of July 2006 hardware for T2s.
- ongoing
197.2: SL to review TC's document on "How CERN helps the small sites to
install and manage the LCG software"
- ongoing
|
