Meeting Notes for 3 March

Cosmic

• Replacement of blessing files:
  • This process has been completed. There are backups of everything in case there are future questions.
• Documentation:
  • Will upload the details for the replacement of the blessing files.
• Bug in ThresholdTimes:
  • All the files have been recreated. Bob and Edit ran some tests to see how they affected the analysis results by running analysis before and after the file switch for a number of splits.
  • Edit will run some parsing comparing old vs new to determine if the julian day is correct when it rolls over.
  • If tests are OK, Edit will move the files to production.
  • Tom mentioned that we still need to keep working to determine the correct calculation of the cpldfrequency at split time. We might need to recreate all these files again in the near future.
• Dynamic plots (Flot vs Plot.ly):
  • We will all meet end of March (24th - 26th) to review the dynamic plots.

CMS

• CIMA:
  • There has been some correction to this code by Michael Soiron that could have improved the performance by closing the connections after they had been open.
• Event-display:
  • TomM will change the unzipping of files at the server by unzipping them at the client. This can improve performance and help prevent server slowness.
• Stress Test before the masterclass:
  • Ken scheduled a stress test for tomorrow Thursday at 1:15pm (for about 15 minutes). We will be using CIMA and at the same time Mihael and Edit will be watching the servers to see if there is any clue as to what is causing the slowness.
• Discussed different alternatives if these two modules keep causing slowness:
  • Move CIMA to CERN?
  • Ken has different options for when the actual masterclasses are happening so the mentors can work with other resources such as loading the event-display from a DVD locally rather than using the server.