LCG Management Board

Date/Time:

Tuesday 5 December 2006 at 16:00

Agenda:

http://indico.cern.ch/conferenceDisplay.py?confId=a063277

Members:

http://lcg.web.cern.ch/LCG/Boards/MB/mb-members.html

 

(Version 1 - 9.12.2006)

Participants:

A.Aimar (notes), D.Barberis, S.Belforte, L.Betev, I.Bird, N.Brook, K.Bos, T.Cass, Ph.Charpentier, L.Dell’Agnello, C.Eck, M.Ernst, I.Fisk, B.Gibbard, J.Gordon (chair), C.Grandi, M.Lamanna, E.Laure, J.Knobloch, H.Marten, P.Mato, P.McBride, G.Merino, R.Pordes, H.Renshall, J.Shiers, O.Smirnova, J.Templon

Action List

https://twiki.cern.ch/twiki/bin/view/LCG/MbActionList

Next Meeting:

Tuesday 12 December 16:00-17:00 - Phone Meeting

1.      Minutes and Matters arising (minutes)

 

1.1         Minutes of Previous Meeting

The comments on the previous minutes (by J.Templon and J.Gordon) regarded the definition of “July 2007” as global target date vs. “May 2007” as proposed at the MB meetings and reported in the minutes. The milestones are still going to be discussed in this and future meetings by the MB and the global target date will be defined then.

 

Minutes approved.

 

2.      Action List Review (list of actions)

Actions that are late are highlighted in RED.

 

  • 21 Nov 06 - F.Carminati agreed to prepare and propose (mail to the MB) the mandate, participation and goals of the working group on “Caching and Data Access”.

 

Done. .Carminati distributed the initial proposal.

 

  • 21 Nov 06 - I.Bird will prepare the mandate, participation and goals of the working groups on “Monitoring Tools” and “System Analysis”.

 

Ongoing. Discussing the mandate with J.Casey and I.Neilson, proposed as joint chairs.

For the System Analysis group I.Bird discussed with J.Andreeva, the proposed chair.

 

  • 25 Nov 06 - Sites should send to H.Renshall their procurement plans by end of next week.

 

Not done. Several sites reiterated their request to “first receive the summary of the requirements from H.Renshall and then provide the procurement plans”.

 

  • 28 November 2007 - H.Renshall will send the updated table of the experiments requirements.

 

Done. First version distributed, plans from the sites expected.

 

  • 29 Nov 2006 - L.Robertson and F.Carminati will discuss with K.Bos about changes to the mandate of the Storage Classes working group.

 

On the way.

K.Bos explained that the Storage Classes working groups is addressing similar questions and the mandate will be expanded and discussed at the GDB.

 

  • 29 Nov 2006 – The MB member send feedback to A.Aimar on the Targets and Milestones for 2007.

 

Done.  The Milestones will be discussed further in the ECM meeting but initial feedback was received.

 

 

3.      Proposal and Discussion on Targets and Milestones for 2007 (Slides) - A.Aimar

 

A.Aimar presented an updated version of the proposal for target and milestones for 2007.

The changes, following comments received after the previous MB meeting (minutes 21.11.2006) are highlighted in RED in the presentation (slides).

 

The goal of the presentation was to foster a discussion on the details of the targets and milestones but the discussion summarized concluded that they should be discussed in the context of the experiments plans at the LCG Experiments Coordination Meetings (ECM agendas).

 

The goal of these “level 1” targets is also to have them as “level 2” targets in the individual plans (for sites, experiments and the projects).

 

Slide 5, not present in the previous version, tries to visualise the high-level general progress in a concise manner. It shows the matrix that could be used to track the progress of the targets (rows) at the sites (columns). In green would be highlighted the site targets (cell) once they are reached.

The cells of the matrix would contain the values to reach and the completion date, once the target is fulfilled.

One can also easily see when 1+8 sites reach a target.

 

Ph.Charpentier noted that would also be useful to see the correlations between different targets and how they affect the real use by the experiments, and not only the decomposition in individual ones.

He also pointed out that for a specific experiment the metric “1+8 sites meeting the target” is not sufficient, if the two missing sites affect heavily one experiment the target cannot be considered achieved.

 

J.Gordon mentioned that procurement at the sites will obviously proceed separately but should be always above that the experiments require by a given date.

 

J.Templon asked whether the values in H.Renshall’s table, about experiments requirements for 2007Q1 and 2007Q2, should be procured at 100% of the values.

A.Aimar clarified that the percentages (e.g. 50, 70, 80, etc) in the slides refer to the value needed to reach by 2008. The values in the Experiments requirements are part of the ramp-up and are what the experiments need by Q1 and Q2 (and to be fulfilled at 100%).

 

T.Cass noted that the activities in 2007 will be driven by the experiments and therefore what happens if the experiments are not sending data at those peak rates because have other activities planned? J.Gordon said that during CSA06 the data buffers were left to fill up by stopping transfers to RAL and this would force rates to rise at peak levels.

 

I.Fisk mentioned that the sites usually can just try to go “as fast as possible” and then see if that reaches the expected peak rate.

 

J.Gordon expressed the worry that the targets are too high for summer 2007.

 

S.Belforte noted that is useful to have targets for the specific components at the sites (e.g. the MSS tapes infrastructure) and others that test several components in the data chain (e.g. DAQ to Tier-0 to Tier-1s).

And also should the sites try some of the targets separately? Or simultaneously as will happen when real data will arrive? The experiments should drive the tests that they consider more relevant.

 

K.Bos stated that we should check whether these targets match the activities that the experiments are planning in 2007. Several MB members agreed that this should be verified during the activities of the experiments and not as ad hoc tests. T.Cass noted that some targets reached in the past with “ad hoc” setups in 2006 could not be repeated easily after a few months.

 

Ph.Charpentier mentioned that the milestones should not be just reached but sustained after they are achieved.

 

J.Shiers said that, because the LCG is now a production infrastructure, the targets MUST be reached within the framework of the experiments activities. The targets should not be reached in isolation because this would not be useful to the experiments and actually would spend resources needed by the VOs.

 

In conclusion the MB reached a consensus on:

 

-          The targets proposed (and/or others) should be taken into account by the experiments into their planning for 2007. But the whole activity should be driven by the experiments and planned in the LCG ECM meetings that are the proper place.

 

-          Experiments will add other new milestones relevant to measure for their activities and targets expected.

 

-          Usually the ECM deals with mid-term issues, but a longer-term view will be prepared in the next couple of weeks.

 

New Action:

19 December - J.Shiers and H.Renshall will report on the progress on the definition of targets and milestones for 2007 at the LCG ECM meeting.

 

1.      AOB

 

 

No AOB.

 

2.      Summary of New Actions 

 

 

19 December - J.Shiers and H.Renshall will report on the progress on the definition of targets and milestones for 2007 at the LCG ECM meeting.

 

 

The full Action List, current and past items, will be in this wiki page before next MB meeting.