October 10, 2018 online meeting minutes

From CLONWiki
Revision as of 09:40, 10 October 2018 by Boiarino (talk | contribs) (Created page with "present: Sergey Boyarinov,, Stepan, vpk, Ben, Graham, Eugene, Raffaella, Nathan, William 1. CLAS12 DAQ/Trigger status * hardware status: everything installed and operational...")
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigation Jump to search

present: Sergey Boyarinov,, Stepan, vpk, Ben, Graham, Eugene, Raffaella, Nathan, William

1. CLAS12 DAQ/Trigger status

  • hardware status: everything installed and operational (BAND coming), will be good to get some more spare boards (TI, VTP etc); remaining problems: occasional TI failings (could be related to fibers or/and something else), occasional trigger links failings; may remove TDCs from some detectors
  • software status: everything installed and operational, recent additions were compression in FADCs and MMs and readout windows shrinking; remaining issues: some run startup and ending failures (can be related to hardware but need better diagnostics); iocscaler replace with amp
  • daq settings: operational; remaining issues: need to cleanup configuration files sequence
  • trigger settings: operational, recent additions are drift chamber trigger (included in all 3 major triggers) and better trigger timing alignment allowing readout windows shrinking; remaining issues: some more timing checks and tuning, like muon trigger delay; new trigger configurations will be implemented on demand (like 2 tracks in one sector, 2 clusters in single sector etc)
  • messaging/rcdb/etc: activemq works in general but seems reaching some limits in the number of messages, effecting rcdb update, rcdb-related messages were moved to separate topic, will continue to optimize topic structure
  • data files recording and transfer to scicomp: subdirectories implemented on scicomp request, data writing and reading using the same partition - works fine with reduced data rate, all DAQ now on one clondaq6 server and we do not need to switch machines; data rate to scicomp still 1.5GByte/sec as at spring time, never entered trottling regime yet, will watch it during production
  • observed crashes and debugging
  • etc