OWG Meeting 5-Dec-2012
- Still need to arrange for Ed to talk more on FPGA programming
- Review of minutes from 7-Nov-2012 meeting
- Monitoring farm, ET systems, simulated raw data, histogram creation, etc. - Elliott, Dave
- EVIO - Carl, Elliott
- stream parser functions
- Counting House status - Hovanes, Elliott
- EPICS - Hovanes
- Controls - Elliott
- Solenoid test status
- JInventory database - Elliott
- Hall A interested
- DAQ group projects- Bryan, Dave A, Graham, Carl, Vardan, Elliott
- CODA3, EVIO, 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]]
- Calibration/Alignment issues?
1:30 PM Wed 5-Dec-2012 CC F326
1:30 PM Wed 19-Dec-2012 CC F326
Present: Elliott W, Simon T, Dave L, Hovanes E, Beni Z, Carl T, Bryan M, Mark I.
See Elliott's slides for complete details. The upshot is that everything is going smoothly concerning installing and running prototype monitoring farm software on the borrowed nodes (4 x 16 cores) in the Counting House. Performance is as expected, network bandwidth limits are easily achieved over Ethernet, cpu loads are as expected, ET is working fine, everything is scaling properly. Dave A is setting up the InfiniBand switch and Elliott will repeat all tests when it is working. Dave L is close to having a JANA monitoring program to run instead of Elliott's test consumers. Dave and Elliott will need to implement the translation table in JANA to complete performance testing. Elliott and Vardan will create a farm manager CODA component when Vardan is done with other higher-priority tasks.
Carl noted that ET will be need to be retuned for IP over IB, as the default parameters are tuned for IP over Ethernet. It is not clear what fraction of the IB bandwidth will be achieved in practice.
Carl and Elliott reported they are working on speeding up buffer i/o, adding stream-parsing functions when full deserialization is not needed, and on adding an ET channel to the evio channel interface.
Hovanes reported on discussions he and Elliott had with Dave Fazenbacker (Facilities) concerning outfitting the control room and the conference room (see the two pictures above for likely designs for each). Someone noted not to forget to install a polycom system in the meeting area.
Hovanes reported he is working on reading out FADC scalers into EPICS, that getting data from the solenoid controls PXI system into EPICS is going well, and that Beni is working on a CSS gui for controlling the FDC high voltage system. Beni would like to implement dynamic channel definitions so he wouldn't have to reprogram the gui for every small change in the HV configuration.
Elliott reported that things are going well, final connections are being made, parts of the system have been leak checked and some are being pumped down, the control system is in good shape, and that things look good for beginning cooldown in two weeks. Still there's plenty to do, including finishing the OSP, but there are no show-stoppers. After cooldown commences we'll need to finish work on the power system and get an OSP for powering the magnet. There will be an review of powered operations in early Jan.
Elliott reported that Hall A is interested in implementing a version, and that Graham mentioned he might want to make JInventory lab-supported in some wat to be determined. Sergey B is still working on the cable component. Elliott hasn't worked on this in a while due to other priorities.
Bryan reported that the first delivery of new rocs will arrive in early Jan (six he thinks for us), second delivery in May. He's setting up a system to perform full crate tests of the FADC250's, after the individual boards are tested in UMass.