WLCG Tier2 Tutorials & Experiment Sessions

15 Jun 2006, 09:00 → 16 Jun 2006, 18:00 Europe/Zurich

160/1-009 (CERN)

ROOM ALLOCATIONS

160 1-009: M/W services: overviews in morning; hands-on in afternoon

28 R-015: 'hot meetings' / break-out sessions / ad-hoc working groups (no agenda!)

40-SS-C01: Thursday morning - ATLAS
40-SS-C01: Thursday afternoon - CMS

40-SS-C01: Friday morning - ALICE
40-SS-C01: Friday afternoon - Applications Area talks

40-SS-D01: Friday (all day) Quattor tutorial

IT Amphitheatre (Tuesday morning): GridPP sessions

Thursday 15 June

09:00 → 12:00 Data Management Tutorials

This session will provide overviews of the various key data management components that are either deployed directly at Tier2 sites, or else provide services to Tier2 sites.

09:00 FTS tutorial

The File Transfer Service is used to download data to Tier2s (e.g. analysis data, datasets to produce calibrations & alignments) and to upload data from Tier2s to Tier1s (MC production)

Speakers: Gavin McCance (CERN), Paolo Badino (CERN)

Slides FTS_Administration_for_Tier-2s.pdf Slides

09:45 LFC tutorial

The LCG File Catalog is required as a local catalog at all sites by ALICE. It is provided at the T0 and all T1s - the latter serving also T2s - for ATLAS. An LFC instance at CERN serves as a CMS global DLS(=DataLocationSystem) catalog. It is provided as a central catalog at the T0 for LHCb.

Speakers: Jean-Philippe Baud (CERN), Sophie Lemaitre (CERN)

Slides LFC-Tier2-tutorial-15_06_2006.pdf LFC-Tier2-tutorial-15_06_2006.ppt

10:15 break

10:30 DPM tutorial

Speakers: Jean-Philippe Baud (CERN), Sophie Lemaitre (CERN)

Slides DPM-Tier2-tutorial-15_06_2006.pdf DPM-Tier2-tutorial-15_06_2006.ppt

14:00 → 18:00 Data Management Tutorials

14:00 xrootd tutorial

Slides XrootdTutorial.pdf XrootdTutorial.ppt

15:30 break

15:45 dCache tutorial

Slides greig-cowan-dcache-tutorial.pdf Lcg-tier2-dcache-june-2006-workshop.pdf

Friday 16 June

09:00 → 13:00 Information System, Workload Management System and Monitoring

Suggestions include:

• on the Information system (sites misconfigured can be black hole ...)
• monitoring and accounting (How we will measure the production at one site ?)
• Something on the submission system of the experiments (DIRAC, BOSS, ProdSyss? Alien ??...to know what is the flow of the jobs, there requirements...) (presumably part of the experiment sessions).
• xrootd if it is required? (See Data Management Tutorials)
• VObox if it is required? (As part of experiment sessions)

09:00 Information System

Slides T2-Tutorial-Info.pdf T2-Tutorial-Info.ppt

09:45 Workload Management System

Slides T2-Tutorial-WMS.pdf T2-Tutorial-WMS.ppt

10:45 break

11:00 Monitoring

Slides monitoring-for-T2-admins.pdf monitoring-for-T2-admins.ppt

14:00 → 16:00 Building High Availability / High Reliability Services

This session will discuss the various techniques that have been used at CERN and elsewhere to address the WLCG Memorandum of Understanding (MoU) targets in terms of service level and availability.

14:00 WLCG MoU Service availability Targets, Monitoring and Display

This talk will give an overview of the currently agreed Tier 0, Tier 1 and Tier 2 service availability targets and the SAM (Service Availability Monitoring) system, based on Site Functional Tests, used to monitor sites.

Speaker: Harry Renshall (CERN/IT/GD)

Slides hrr.pdf WLCG Service Availability

14:20 Technologies to Implement High Availability Grid Services

A review of technologies for High Availability of Grid Services and how they can be implemented. This will include hardware hardening, high availability linux and DNS Load Balancing.

Speaker: Tim Bell (CERN/IT/FIO)

Slides Technologies to Implement High Availability Grid Services Tier2_Availability.pdf

14:50 CERN local High Availability solutions and experiences

The current CERN deployments of high availablity grid services and experiences with them will be described.

Speaker: Thorsten Kleinwort (CERN/IT/FIO)

Slides GridHW.pdf Slides

15:10 CERN Physics Database Services

This talk will cover the Physics Database Service organization and expected evolution in terms of hardware solutions and service structure

Speaker: Maria Girone (CERN/IT/PSS)

Slides mg-t2workshop.pdf Slides

15:35 High Availability Databases based on RAC/Linux

This talk will address the hardware and software redundancies for HA services and summarize the first experiences of one year in production.

Speaker: Luca Canali (CERN/IT/PSS)

Slides High availability databases based on RAC/Linux T2_tutorials_Jun06_OracleRAC_LC.pdf