2008/01/09 CSEP meeting

Attendees: Tom, Phil, Danijel, Jeremy, John, David, Yan, Mansi, Amit, Masha

1. David Meyers presented JIRA issue tracking software package as alternative to the Trac package CSEP is using.
2. Danijel brought up discussion of the CSEP abstracts for the SSA that are due this Friday:
   • Tom suggested to have an abstract for oral presentation - Danijel
   • Phil suggested to have a poster about testing center development - Masha
   • David suggested to outline models that are ready to be incorporated into the testing center - to invite people, to stimulate more models to be entered into the center
3. Status of CSEP V8.1 release:
   • Installed on csep-op on Thursday, January 3, 2008
   • Phil asked about new features of the release:
     1. No random numbers stored to the files
     2. Ability to reproduce failed test dates
     3. ANSS downloaded catalog is filtered to include EQ events only
     4. Distribution does not include installed forecast models, acceptance tests are not affected by installed models
     5. Installed 7 3-months models from David Rhoades
        • Forecasts for the testing period of 12/1/07 - 3/1/08 were generated on Friday, January 4, 2008
   • Masha mentioned that processing time doubled up with use of new random number generator
   • Tom asked about processing time scalability with more models in the center:
     1. N and L test per each model
     2. R-test per every possible pair of models
   • Tom and David discussed the ways of improving R-test
   • Tom asked about stability of network:
     1. John is working on getting servers set up on their own network, so processing is not affected by outside network going down
4. Scheduled tasks for the next release of CSEP V8.4:
   • There is a number of known problems that would need to be addressed:

     1. Trac ticket #60 Lowering minimum magnitude for one-day models input catalog causes Matlab recursion limit error in STEP

     2. Trac ticket #61 Importing of over 1 million of events into Matlab code never finished after 20 hours of running

     3. Trac ticket #62 File IO used by random numbers doubles up processing time

     4. Trac ticket #36 Re-implement evaluation tests without random numbers

        • Jeremy suggested not to wait for the write up paper from David Rhoades, and start working on at least N-test - Jeremy will help
   • As Danijel suggested in one of the previous meetings:
     1. We should work on web interface for the CSEP results.
     2. Consider implementing storage API - "database"-like interface to query existing operational data
   • Phil suggested to put together an interface for the data access
   • Yan asked Masha to visit UCLA to start work on Western Pasific forecast
     1. Phil asked about catalog data used by the forecast - CMT
   • Yan will set up schedule for global testing - to be entered into the testing center later