Choose timezone
Your profile timezone:
PASTE meeting
→
Europe/Zurich
2-R-014 (CERN)
2-R-014
CERN
Description
CERN, bat 2 R-014
To join the telephone conference, click here: https://audioconf.cern.ch/call/0165202
To dial in to the conference:
a. Dial +41 22 7676000
b. Enter access code 0165202
Production and PASTE Meeting, 6th June 2010, CERN
Chair: A.Tsaregorodtsev
Minutes: F.Stagni
CERN: J.Closier, Z.Mathe, R.Santinelli, F.Stagni, A.C.Smith, S.Paterson, P.Charpentier, S.Pedrazzini
Phone: A.Tsaregorodtsev, A.Zheleyov
0. Ongoing Action List
Updated in Savannah task list.
1. Production Issues
There is still the long ongoing issue about the LSF time utility. The meeting in London this week maybe will help in defining a more definite solution.
Request page: the possibility to validate a prodution request has been discussed. Stuart commented that anyway a techinal expert should anyway have a look at that, so a test would be still a partial one, and there would probably be no possibility to completely automate the production. Anyway, a solution is envisaged, but it is a long one, and should involve not only DIRAC or bookeeping developers. After a comment from Gloria, and some following discussions, it has been agreed that Zoltan will provide a proposal and attach it to the task.
At the moment, no productions are running. We are stuck, waiting for a new version of DaVinci that will reduce the time to process an event by at least a factor 6. The ball is now on the physics community. Advancements in Brunel are instead much more difficult, so for that we'll have to wait for a longer time. With a rough estimation, a 1M HS06 for job will be enough.
The only DIRAC part where we should not fail is about pointing to the right queue.
2. DIRAC Release Update
Three patch releases have been done while in St.Petersburg to deal with the developments with LSF time utility.
3. DIRAC Tasks
Nothing to report that has not already updated in Savannah.
4. AOB
3 new machines have been requested, one at least will need to be used for the accounting.
CHEP abstracts: we still don't know if Andrew's paper has been accepted. Anyway, Philippe and Federico are going, and Andrew probably will.
LHCb Jamboree in London: a slot of 2 hours have been allocated on Thursday morning, not in parallel with the other meetings.
- Stability of SEs: where are we?
Andrew is not going to London, he said there's nothing very new to present. This could be put in AOB, to see if sites have something to update. Anyway, concerns that we had and have should be reiterated, so Andrew will prepare a couple of slides to trigger the discussion.
- CPU normalisation
Ricardo can and should present something on this topic.
- Warning on possible effects of large pile-up at LHC: increase in queue length, additional resources?
- Wednesday afternoon: our updated plans for testing "dynamic data placement". Follow up from Amsterdam latest discussions. Idea is to use the sDST and DST as caches. Plans was made for using p2p and commercial softwares. Tests could be made in DIRAC without many modifications.
There are minutes attached to this event.
Show them.