vdas_20111020
1. Announcements
2. Software Engineering Run 1 preparation: projects + organization
- data production (ER1 Virgo mock data: Benoit, Didier)
- data storage (where? how much? disk space?)
- low latency transfer (where?)
- DQ information (which one? simulation)
- Search pipelines (which one? where are they running? which DQ information?)
3. Current issues:
- Data transfer to computing centers status (Livio, MAB)
- VDB (Gary, Leone, Didier) [VDB roadmap]
- CC data access
4. AOB:
- Resume of the DASWG F2F workshop
- Computing needs for 2012: JECC documents
Attendees: Nicolas, Florent, Elena, Cristiano, Alberto, Didier, Benoit, MAB (taking minutes)
Minutes:
MAB:
Nicolas has been proposed to be the Software ER coordinator for Virgo. Should be appointed at the next VSC. Agreed to discuss SER matters in VDAS call.
Goal of the SERs: prepare advanced detectors data analysis where online searches will run from day one.
Goal of the 1st SERs: put in place the first layers of a LV online searches: data production, data transfer, data quality and run few pipelines. Simulated data. No real data access.
Didier presents plans about data production: playback VSR4 data with GPS changes. generate h(t) frames from them including DQ information. All this will use the present DAQ process. See scheme. VSR4 playback data won't be saved.
Question: Gaussian noise or MockData real data?
Benoit: not completly define but we are ready to switch from one option to the other.
Question: do we provide a combined cat2 DQ info or not.
Florent: each pipeline should be able to apply the DQ that is really needed.
Answers: many pipelines want to apply blindy DQ info. Need to provide this, but things will involve hopefully but for SR1, a combined info should be OK.
Low data transfer: use of shared memory in Virgo and LIGO. Still discussions with LIGO. Virgo is ready. Transfer to 1 or the LIGO sites (CIT, LLO or LHO). And Cascina will receive L1 and H1 data.
Bulk transfer: use of LDR. Should be easy to do, but need to check the status of LDR (upgrade in january?). Need to transfer data to CC? Need to check with Livio.
Pipeline: MBTA (few changes wrt VSR3 version).
Elena: NOEMI and all related monitors will run on these data at Cascina. want also to run NOEMI on LIGO data.
MAB: may be too early to try to do that. No LIGO RDS will be generated. No interest to run NOMEI on simulated data. Could require useless work. Need to understand better LIGO plans about real data production. A next VDAS will be dedicated to software for advanced Virgo.
Teams:
- data production (Benoit+Didier)
- low data transfer (Benoit+Didier)
- bulk data transfer (Livio+MAB)
- data quality (Didier, + ....)
- pipelines: MBTA (xxx?), NOEMI (Elena, Alberto), what else????
Nicolas will now take care of following up these projects
3. Current issues:
VA3 data transfer: MAB is working on this. Hopefully finished in 1 week.
data access in Lyon: MAB is working on this. Problem was supposed to be fixed. New problem?
JECC meeting: computing requests. Issues with matlab licences at CNAF. Requests for CNAF space disk next week.
Elena will provide request for NOEMI in 2012.