[Halld-online] Online Meeting minutes 9/10/2014

[David Lawrence]

Hi All,

  Minutes from today’s online meeting are no posted on the meeting page and copied below. Please make any corrections to the wiki page.

https://halldweb1.jlab.org/wiki/index.php/OWG_Meeting_10-Sep-2014

Regards,
-David


Attendees: Alex S., Bryan M., David L., Simon T., Sergey F., Curtis M., Hovanes E., Dave A., Sergeui P., Beni Z.
[edit]Announcements

There was a power outage in the counting house today. The power is now back on. Cause is unknown. It is also unknown which if any gluon machines were affected since they should mostly be on UPS.
[edit]Trigger

Reinstalled 22 CTP boards after replacing thermal pads
Still have some funny issues with temperature monitoring on a couple of the FPGAs
Chips did not seem too warm to the touch so we don't think anything is being damaged
Currently have 2 algorithms running on GTP
BCAL north cosmic trigger has been tested and is working
Ran at ~200Hz for >20hrs
Next step is to get full BCAL trigger working. After that, TD's will all be moved into a single TS crate
Trigger commissioning plan is under development
Will come after the photon beam is established in the overall commissioning plan.
Tagger microscope will be moving into Tagger Hall next Friday.
Do not plan to have tagger electronics in hall during commissioning.
[edit]DAQ

Still waiting to test full system (need all TD's in single TS crate)
Currently expect to run up to 8kHz and testing so far indicates this should not be a problem
Preparing OpenSUSE node (gluon49) to test newest Linux kernel for DAQ system crashes
This will likely be done during the large scale DAQ system testing tomorrow
CODA 3.02 is n final testing stages. Could be ready for use in Hall-D as early as tomorrow
Will also be installed in /site once it is ready
Potential performance improvements are likely to push up the current 8kHz limit
Several bug fixes including the one that causes crashes in the secondary ROL at end of run
A graphical monitoring program "daqmon" has been developed
Presents histograms generated on ROCs
Uses dedicated communication mechanism in addition to RootSpy's cMsg based system
Allows for faster updates for some complex displays where greater than 1Hz update rate is desired
[edit]Monitoring

Start.stop scripts are now ready and need testing with DAQ system
Starts/Kills monitoring processes on multiple remote computers
Supports multiple monitoring levels (occupancy, reconstruction,...)
GUI has been built that presents integrated rates and lists active nodes.
Can be used to start and stop monitoring processes, but usually that should be done by DAQ system
[edit]Controls

Vanik and Nerses finished with the Voltage Control Apps.
Not all of them have been tested by the detector groups yet
Next focus is on the motor apps which will include the harp scan software
Work has started on readout of 2 Beam Position Monitors
First attempt tried to port existing software from VxWorks to Linux. This proved more complex than first anticipated
Reverted to using VxWorks where driver for Struck VME scalers already exist.
Hovanes has been working with Mark Ito on the alarm system.
Alarm system configuration can be dumped into an XML file and read back in
[edit]DAQ Config. Parameters in Data Stream

David presented a scheme for implementing some DAQ configuration parameters into the data stream
System defines an EVIO bank that will be written out by each crate for each DAQ event
Initial estimate is roughly 1kB per DAQ event or about a 5% increase for a 19.4kB event size
If in multi-event block mode, percentage decreases (should be O(0.1%) for high rate running)

-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://mailman.jlab.org/pipermail/halld-online/attachments/20140910/6ab3c70d/attachment.html