Attendees: Daniel, Serge, Bill, Douglas, Jacek

integrating time series analysis into qserv etc
  https://dev.lsstcorp.org/trac/wiki/dbTimeSeries
  - need separate system alongside qserv to run
    non-sql "queries", indexes will be in specialized
    format
  - consider using hadoop
  - need to think through how to integrate w/qserv,
    eg for queries that need specialized time series
    analysis that are based on object ids or have
    spatial constraints
    - could be done in user code,
    --> need dedicated meeting to discuss details


reworking scripts that partition csv input data
  - want to throw arbitrary number of resources at
    partitioning
  - implemented such script, almost finished
    - script is using local mysqld instance on each
      batch machine as a "scratch" space
  - almost done with (re-)partitioning full W13
    - also packed the data using myisampack
  --> prepare space for dumping qserv-ready
      on lsst10 [Jacek]


concurrency test on yilis
  - Daniel to email queries to Serge


Bill
  - tried to simulate different error conditions
  - ready to merge with master
  - next step: fault tolerance (trigger error during
    query execution, catch it, rerun query under the
    hood, all without user knowing about the extra
    activities)


parser
  - fixing near neighbor


push metadata to master?
  - it will be disruptive (now we need to run qms,
    metadata must be loaded)
  - warn the whole team first, especially the in2p3 group


no meeting next week
  - Jacek @mysql conf, Daniel taking ~2 days off

Jacek

########################################################################
Use REPLY-ALL to reply to list

To unsubscribe from the QSERV-L list, click the following link:
https://listserv.slac.stanford.edu/cgi-bin/wa?SUBED1=QSERV-L&A=1