Meeting on VESF DA School, April 13th, 2011

Outcome (selection) of meeting with Antonella Bozzi last week

• Antonella is going to create a dedicated “vesf” web password for accessing protected pages. Please let her know which pages should be accessible.
• Antonella proposes farmn machines for the school. They are directly accessible in the network, run SL5.3 at 32bit.
• farm1-7 are monoprocessors, 4 cores, 2 GByte ram, the 8 and 9 are biprocessors, 8 core overall, 4 GByte RAM.
• All run Matlab
• No problem to have olserver14 and 17 availalble as backup, as discussed. They are SL 4.2
• Help by Livio offered for data transfer. 
• No issues about setting up students Unix account.
  

 Important: do not count on full support at EGO during the Easter holidays, which run more or less from April 21st to April 26th inclusive (school break ...)

Software

Mail from G.Di Biase

     PCs dedicated to VESF School 2011 are FARMNx.virgo.infn.it. These machines are SL 5.3 i386. Following standard installation for this release (SL53) I can install following version of software indicated:

ds9-6.2

funtools.i386 0:1.4.4

sextractor.i386 0:2.5.0-6.el5

match-0.11

octave-2-9-10 (NEED 3.2.4 at least. Straightforward to install. Compiling from source is simple.)

Test using farmn1.virgo.infn.it . Now only this machine has all previous software installed.

I also remind that all FARMNx have matlab 7.11 ( /opt/sw/matlab711/bin/matlab )installed and usable with no license limit.

• Other software on farmn? Software in the Virgo environment is sufficient
  

• No need to run KleineWelle or Omega: triggers will be ready for students to play with
• Need to run OmegaScans already performed on loudest triggers.
• BRMSmon to be run by students.
  
  

Data

• Severe issues while retrieving data from Lyon. Timeouts accessing the files on tapes. Scripts stop.

• Nicolas provide the scripts to Livio and Elena who will carry out the job.

• Amount of data: some more than 1 TB, need two buffers (bufferv30 and bufferv31)
  

Mail from Didier
On request of Elena, I send again (see below) the sciencemode segments that I selected.
They determine the two weeks of VSR1 data (Aug10,07-02:44:22 to Aug27,0701:13:48) that we will take back to Cascina's buffers bufferv30 and bufferv31.

Nicolas made a selection of 160 channels to do a RDS which represents about 2 TB of data for those two weeks. We will try to reduce the number of channels so that extracting those two weeks from Lyon's tapes, producing the RDS and transferring data to Cascina does not take too much time.

Meanwhile, VSR1 h values can be used. I just checked that they are still in Cascina, in bufferv7 and bufferv11. They are available through the ffl /virgoData/ffl/VSR1/HrecV3.ffl

 

Data Quality

Mail from Didier

Here are some information I would like to share before the meeting.
For Data Quality, I plan to provide to the students only scripts and ffls:

- /virgoData/ffl/vesf_rawdata.ffl : pointing to the VSR1 raw data RDS

- /virgoData/ffl/vesf_h.ffl : a selection of VSR1/Hrec-V3.ffl for only the chosen playground period.

- /data/procdata/bufferv3/dq_vesf : contains all the scripts the students will use. Those scripts will call the various tools used to produce DQ flags or estimate their performances.

More details are in /data/procdata/bufferv3/dq_vesf/run_dq.sh
which shows a draft version of the procedure I will give to the students for the practical session.

Other tasks

• Ask Antonella about disk space. Home, and what more?
• Ask Antonella to make available one typical student account for testing. For instance, create a dummy account, from which students are able to copy scripts and anything to be shared, and to be tailored and propagated to other accounts.
• Prefer wired connection.
• Start defining the schedule, taking into account catering constraints.
• Next week meeting! Same time, with accounts and software setup… hopefully