OWG Meeting 19-Jun-2013
- Python/EVIO for CLAS12
- What is a good data file size?
- Review of minutes from 22-May-2013 meeting
- C++11 Usage in CLAS12 (example)- Johann G.
- RootSpy progress - Sean
- Role of JInventory database - All
- cables, translation table, HV/LV, etc.
- Do we need a separate conditions database - All
- i.e. will the CCDB suffice?
- if not, when do we need a prototype conDB?
- Status of DAQ testing in Rack Room - Dave A
- Counting House update - Hovanes, Elliott
- Solenoid test update - Elliott
- EPICS update - Hovanes
- Controls update - Elliott, Hovanes
- DAQ group projects- Bryan, Dave A, Graham, Carl, Vardan, Elliott
- EVIO, cMsg, CODA3, ECS/CSS, disentangler, readout list, Java/cMsg GUI
- Electronics, trigger and DAQ status reports - Chris, Alex, Dave, Fernando, Ben, Ed, William, Bryan
- Mantis Task Tracker
- Midas Elog
- Upcoming topics
- C++11 in Hall D online - Elliott
- L3 program - Dave, Elliott
- Translation table
- raw data, compressed vs uncompressed
- Calibration/Alignment issues?
- Status of processing uncompressed and compressed simulated raw data and monitoring/L3 farm - Dave L, Elliott
- What online projects can other groups take over? - Elliott
- Status and schedule of Hall D projects proposed to be done by DAQ group - Graham
- Linux SBC configuration, installation, booting - by Bryan
- InfiniBand hardware specification, purchase, installation - by Dave A
- Farm manager CODA component - by Vardan
- Ed to talk more on FPGA programming
1:30 PM Wed 19-Jun-2013 CC A110
1:30 PM Wed 3-Jul-2013 CC F326
Present: Elliott W, Mark I, Nerses G, Hovanes E, Dave L, Simon T, Johann G, Serguei P, Beni Z, Sean D, Curtis M, Justin S.
C++11 in CLAS12
Johann G (from CLAS) gave a talk on C++11 features he, Sebastian M and others found useful for CLAS12, see his slides and example above. Concerning safety he emphasized use of smart pointers and move semantics. Concerning utility he mentioned use of initializer lists, auto, range-for and the new thread library. They also continue to use the BOOST libraries.
CCDB and Conditions DB
We had a long but inconclusive discussion on whether the CCDB will suffice for online use, or whether we need a lighter-weight conditions DB (conDB). We also discussed the nature of the relation between the conditions database and the trigger configuration database, if any.
- Will Dmitry do this? How much time does Dmitry have?
- Priority compared to trigger database? Who can work on this other than Dmitry?
- Is trigger config database a subset of the conDB?
- Need to sweep EPICS data from archive into CCDB or conDB.
- conDB lower overhead than CCDB.
- Is history needed in conDB?
- What is the efficiency of queries by run number in either?
Sean gave an update, see slides above:
- Hists are being gathered at end-of-run.
- GUI and other improvements in the works.
- Need to test with monitoring plugins.
- Sean will visit JLab in late Aug.
Serguei P presented the current status of the inventory database:
- What role will the db play? Source of translation tables? Trigger info?
- Want to minimize sources of information, avoid duplication.
- Need buy-in by Fernando.
- What role might spreadsheets play?
- He is working on entering electronics info from Fernando's spreadsheets and creating specialized pages for electronics.
Hovanes described recent progress in outfitting the CH. Consoles are installed, no power or telecom yet. Paul will be back next week to install cubicle computers and some servers. PR for remaining equipment in progress.
Elliott reported continuing problems with the cryo system and the Danfysik power supply. Not clear when the next test will start. Committee urged making the PLC-based QD the primary and using the Danfysik as secondary.
- Hovanes reported some problems/limitations with CSS, including installation problems.
- He is currently looking at the alarm system "BEAST" from SNS.
- Planning alarm organization by detector, not by function (e.g. not by HV).
- Planning for paging, authorization and identification. Alarm deferral?
- Nerses got HV/LV info from spreadsheets for CDC.
- IOC code will map geographic and detector names to PV's.
- Crate IP names will be functional, not geographic.