LCG Management Board

Date/Time:

Tuesday 30  May 2006 at 16:00

Agenda:

http://agenda.cern.ch/fullAgenda.php?ida=a061503

Members:

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

 

(Version 1 – 01.06.2006)

Participants:

A.Aimar (notes), D.Barberis, L.Bauerdick, I.Bird, K.Bos, S.Belforte, F.Carminati, T.Cass, Ph.Charpentier, I.Fisk, D.Foster, J.Gordon, B.Gibbard, A.Heiss, E.Laure, M.Lamanna, P.Mato, M.Mazzucato, G.Merino, B.Panzer, Di Qing, L.Robertson (chair), J.Shiers, Y.Schutz, J.Templon

Action List

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

Next Meeting:

Tuesday 6 June 2006 at 16:00

1.      Minutes and Matters arising (minutes)

 

1.1         Minutes of the Previous Meeting

Minutes approved.

1.2         Deployment priorities

During the week it became apparent that there was some confusion over the gLite 3.0 deployment priorities. I.Bird sent an email with the clarification of Glite 3.0 deployment priorities to the GDB mailing list.

1.3         Status of gLite 3.0 deployment

Most sites have installed their services but very few have deployed and configured the FTS services (exceptions RAL, FZK and PIC).

 

Sites having deployment issue should report them via GGUS. Currently there are very few GGUS tickets about gLite 3.0 deployment.

1.4         Tier-2 Sites

The Tier-2 sites should have been ready by the 1 June 2006 but is delayed. The discussion at the LCG Coordination Meeting, of which Tier-2 sites will participate to SC4, is not yet concluded.

 

2.      Action List Review (list of actions)

 

 

  • 20 May 2006 - All SC4 sites send accounting data using the report form that will be sent to them by Fabienne Baud-Lavigne (for Jan, Feb, Mar, Apr 2006).

Done. Accounting Data received. The summary is available and will be discussed at next MB meeting, feedback and suggestions are welcome.

The “Installed capacity” at CNAF is missing. It is needed in order to complete the statistics.

Update:

The accounting data for January to April 2006 is now available in:

http://lcg.web.cern.ch/LCG/MB/accounting/April2006_accounting_summaries.pdf

 

The Excel workbook used to generate the report is also available at the following web address:

http://lcg.web.cern.ch/LCG/MB/accounting/Master%20accounting%20summaries.xls

 

Please send me (or to lcg.office@cern.ch) any corrections and comments.

 

  • 21 May 06 - J.Shiers and M.Schulz: Flavia’s list should be updated, maintained and used to control changes and releases. A fixed URL link should be provided to that list.

This will be clarified after the gLite 3.0 presentation at a forthcoming MB meeting.

  • 23 May 06 - Sites should send to lcg.office@cern.ch the names of the contact persons attending, possibly in person, the Internal Review meeting.

Missing from ASGC and INFN.

  • 23 May 06 - Tier-1 sites should confirm via email to J.Shiers that they have set-up and tested their FTS channels configuration for transfers to Tier-1 and Tier-2 sites.

At the time of the meeting only: RAL, FZK and PIC have done it. The information should be sent to J.Shiers in the framework of the SC4 Coordination Meeting.

 

The key points of the configuration are:
- Publishing the SRM end points in the information system
- Installing and set up the FTS services

 

There is FTS documentation written by G.McCance and the information on how to setup FTS is available since a long time.

 

L.Robertson expressed concern that these services have still not been set up, although the issue has been brought to the attention of the site management at the last two MB meetings, and we are only days from the start of the SC4 service. At next MB meeting we should discuss what the problems are..

  • 26 May 06 - Sites should complete and send the questionnaire for the Internal Review to the chair, V.Guelzow.

J.Templon will send to A.Aimar the list of sites that have (or not) responded to the questionnaire.

 

Update: Questionnaire missing from IN2P3 and INFN.

 

 

3.      SC4 Status: Summary of Experiments Plans for SC4 Start-up (more information) – J.Shiers

The attached draft is based upon input received from the experiments up until May 17th - the target date agreed - and as compiled by Harry Renshall (see link to Wiki and attached documents). Some additional clarifications received after that time by e-mail have also been included.

 

 

The attached draft (SC4-experiment-plans-May30.pdf) is based upon input received from the experiments and summarizes the information on the SC4 Plans wiki page maintained by H. Renshall.

3.1         Dteam

The background FTS transfers will start on June 1st and continue at least until the end of SC4.

 

The focus is to debug site monitoring, operational procedures and ensure that that we have the ability to quickly ramp up to the nominal values. Currently the CERN configuration allows transfers up to 800 MB/s. Therefore it can only be used to test ramp up and stability, not the full nominal performance rates to all sites.

 

The transfers will use the same end-points used during the April SC4 tests. And, once the transfers to disk are satisfying the requirements, a schedule for ramping up the disk-tape transfers to nominal rates will be defined.

 

From June 1st sites (Tier-0 and Tier-1) should operate the required services in production quality mode, with constant service and support.

3.2         ATLAS

The ATLAS SC4 exercise will start on June 19th and run for 3 weeks. All details are here.

The preparation has already started and will ramp up in the coming weeks. Tier-1 sites will have to be all validated, starting on the 1st June. The exact schedule of the sites validation will be distributed. The selection of the disk or tape transfers is defined by different endpoints, depending on the type of data storage.

 

The table on page 2 shows the rates expected by the ATLAS SC4 tests. 40% of these data rates should be written to tape and the rest on disk. The minimum lifetime of the data is of 24 h: after that disk space and tapes can be reused. Tapes should be unmounted before being reused, in order to avoid quick-erase procedures.

 

3.3         CMS

CMS plans to have transfers to temporary disk space of 20 MB/s to/from each Tier-1 site  and run 25,000 jobs/day before the end of June.

The job duration will be tuned in order to reach the target of 25,000 a day, but should be about 10-12 hours

 

In addition to the transfers from each Tier-1 to Tier-1 site, there should also be transfers from Tier-0 to all Tier-1 sites and from all Tier-2 to all Tier-1 sites at same 20 MB/s rate.

 

Currently most transfers are using the standalone PheDex server at the sites. When FTS becomes available on a site PheDex will use FTS for all transfers.

 

The sites already have information on how to set up FTS for CMS. The documentation is on the CMS SC4 wiki pages.

 

Action:

10 Jun 06 - CMS will send to J.Shiers some defined milestones for the CMS SC4 exercises.

3.4         ALICE

In July 2006, for 3 weeks, ALICE will transfer data at 300 MB/s from CERN to a set of Tier-1 sites: CNAF (20%), CCIN2P3 (20%), GridKA (20%), SARA (10%), RAL (10%), US (one centre to be defined) (20%).

 

The goal of the exercise is to test service

-          stability and integration with the ALICE FTD (File Transfer Daemon)

-          7 days of sustained transfer to all T1s.

 

The proposal of Alice for the FTS transfers in Summer 2006 is attached. It also specifies (page 3) the usage of part of the CERN capacity as a Tier-1 site. This usage will be clarified with CERN.

 

3.5         LHCb

The LHCb schedule and plans for 2006  are very clear and are available from the SC4 LHCb plans wiki page.

 

Starting from July LHCb will distribute "raw" data from CERN and store data on tape at each Tier1. CPU resources are required for the reconstruction and stripping of these data, as well as at Tier1s for MC event generation.

3.6         Actions

See all actions listed on the document.

 

The main actions discussed (repeatedly) at the meeting were:

-          Disk and tape end-points must be established and tested for all VOs based on the schedule above. They must be available for testing a minimum of 2 weeks prior to the schedule outline above.

-          All end-points must be published in the information system.

-          All Tier1s must setup and test channels to all other Tier1 sites and their supported Tier2s as described in the following document: https://cern.ch/twiki/pub/LCG/FtsServerInstall15/SC4FTSsetupplan.doc

 

An example Tier-1 site can be found at https://twiki.cern.ch/twiki/bin/view/LCG/FtsServerDeployExampleTier1

 

These actions are managed through the weekly service coordination meeting.To provide MB visibility, this list of actions should be added to, or be reachable from, the SC4 Plans wiki page.

 

Action:

10 Jun 06 – J.Shiers will add an Action List to the SC4 Plans wiki page.

 

 

4.      GLite 3: JRA1 Middleware development plans (transparencies) - J.White

-          Open Issues from Task Forces (more information )

-          LCG SC4 Services: Flavia's List (more information )

 

 

Postponed to next MB meeting.

 

5.      AOB

 

 

No AOB.

 

6.      Summary of New Actions

 

 

Action:

10 Jun 06 - CMS will send to J.Shiers some defined milestones for the CMS SC4 exercises.

 

Action:

10 Jun 06 – J.Shiers will add an Action List to the SC4 Plans wiki page.

 

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