By Anil Kumar
Agenda for the workshop:
http://agenda.cern.ch/fullAgenda.php?ida=a044825
1/25/2005
Flight Departure got delayed at Chicago by 2:00 Hrs.
Currency Exchange Rate: $1 = 1.04 CHF
1/26/2005
Flight Departure at Frankfurt got delayed by 2 hours.
Currency Exchange Rate: $1 = 1.15 CHF
Note: It was better to exchange $ for CHF in Switzerland Vs US
1/27/2005
Attended the Presentation by Dirk Duellmann (CERN) on Overview of C/C++ DB APIs. It explained access to databases using C/C++ as Programming Languages.
Pro*C is not good for dynamic SQL. OCCI is being used to develop POOL/RAL framework.
RAL software is addressing needs for POOL Relational Components, POOL object storage mechanism and ConditionsDB.
RAL is independent of database vendors. Oracle plug-ins work in all cases. MySQL and SQLLite plug-in works in 99% of cases. It is successfully being used in implementation across all POOL application domains.
In CERN Benthic and Tora tools proposed for Interactive Analysis and SQL optimization.
Object Storage into RDBMS using POOL storage mechanism has been achieved using RAL. Technically neural object description in RDBMS is feasible. The RelationalStorageService is by itself a stress test case for the RAL plug-in.
Andrea Valassi highlighted the project goal, manpower available for project and milestones. Project goal is provide common software and tools for non-event time-varying versioned data.
Attended the hands-on session on POOL/RAL. It was good learning experience.
Did some hands on schema creation, Database queries using POOL/RAL APIs against Oracle Database.
Deployment
Presented the Case Study on Database and application Deployment at Fermi Lab
http://www-css.fnal.gov/dsg/external/workshops/lhc_db_workshop_files/frame.htm
There were around 80 attendees. IT/DB at CERN commented that CERN IT/DB would pick up lot of standards mentioned in the presentation. Got the feed back from attendees that it is very impressive Case Study.
Close Out Comments by Maria Girone :
There were more than 80 participants in LHC Database workshop. IT/DB will be happy to organize follow-up workshop. She thanked the participants, speakers, hands-on helper and Jamie Shears and Frank G. for organizing. Then Frank G. thanks IT/DB for all this. It was a great workshop with lot of coverage and participants would be benefited from this exercise.
1/28/2005
Attended the CMS experiment talk by Frank Glege. He explained the on-line and offline databases. Data Distribution from on-line to offline databases. On-line database will keep the minimal data needed to run the detector, but offline will have
Everything. Format of data in offline will be different than on-line. Also offline will keep lot of summary information. Also presented the status and thoughts on data model from Fermi Lab. These slides were prepared by Lee L. of Fermi Lab.
Observation: If format of data on offline database will be very much different than on-line data format, then replication of data from on-line to offline will be complex due to fact that the transformation of data format in replication make the rules of data replication complex and eventually may cause a performance bottleneck in replication. If offline database will have lot of summary information, which is in fact demoralized information, then offline database is considered to be database warehouse. Database warehouse has different design than OLTP kind of databases.
So offline db should be implemented keeping in mind the dataware technologies.
ETL (Extraction, Transformation and Loading) tools should be considered for populating the offline databases Vs replication techniques.
Discussed the application tuning and database tuning concepts implemented with CERN/IT DB
2:00PM-4:00PM Discussion on Data Model Design for CMS
RPC Database.
Met with Archana Sharma - CERN and Marcello Abbrescia –
CERN to discuss the database requirements for CMS Forward
(Endcaps) RPC database. And proposed a draft data model.
And explained the concepts of data modeling so that data
Model can be refined as requirements will be refined.
4:00PM Close out of Experiment Talks
1/29/2005 Week End
Visited CERN Reception and MICROSOM Exhibition Hall of CERN
Guided tour to ATLAS Detector
1/30/2005 Travel to Chicago from Geneva, Switzerland
I found the stay at CERN safe and secure. However, I was bit caution while walking because unexpected/unusual/uncommon snow in the area. Crew was not well prepared to treat the path ways in the campus.