2/16/2015 Notes

DocDB Event


Craig G., Alec, Jan, Zukai, Pengfei, Andrew, and Martin (taking notes)

Memory Usage

  • An event with high multiplicity in the Hough tracker uses up tons of memory due to n^2 nature of the algorithm.
  • We need some protection against this. Perhaps a requirement that the number of hits stay beneath a certain threshold.
  • For now, we will increase the ulimit memory threshold to 8 GB.
  • We should have an error trigger issued for events like these. We can just use the FEB Overflow bit for this sir of stuff.
  • Andrew and Pengfei will have a look at the Hough Tracker code

DAQ Configuration for Streamer Trigger

  • I have implemented the calibration trigger configuration in the same way as for the DDT.
  • Alec and I discussed this a bit and he is willing to change the GT code to deal with a variable number of calibration triggers for the sake of flexibility.
  • This will then allow a user to add several lines to the calibration_trigger table for all the different pulsers they can think of.
  • We got permission to deploy this tomorrow morning, but considering the GT changes required and the fact that the config changes cannot be tested on FCCDAQ right now, I think that we will wait to do this until a later point.
  • I will inform Alec once I have a working version of the new DAQ configuration on novatest01/Partition 2.