OWG Meeting 15-Aug-2012

Agenda

• Announcements
  • Sarin will give another Danfysik tutorial at the Controls meeting on 23-Aug
  • Ben will give a tutorial on FPGA programming on 29-Aug
  • Beni will talk about the gas system at the Controls meeting on 6-Sep
• Review of minutes from 1-Aug-2012 meeting
• Mantis system usage - Elliott
• Computer system manager - Elliott
• FDC test and FADC125 readout problems - Beni, Bryan, Elliott
• JInventory database - Elliott, Hovanes
• Reading test data in JANA - Dave
• Counting House status - Hovanes
• Summer student projects - Dave, Elliott
  • DED
  • RootSpy and mini-Hbook
• Common Logbook System - Elliott
• DAQ group projects- Bryan, Dave A, Graham, Carl, Vardan
  • Raw data tape
  • CODA3, EVIO, ECS/CSS, disentangler, readout list, Java/cMsg GUI
• EPICS - Hovanes
• Controls - Elliott
  • Solenoid test: FastTrack schedule, PXI status, FMEA, QD, vacuum, connection to Cryo group PLC, OSP's, etc.
• Electronics, trigger and DAQ status reports - Chris, Alex, Dave, Fernando, Ben, Ed, William, Bryan
• Mantis Task Tracker
• Midas Elog

• Calibration/Alignment issues?

Time/Location

1:30 PM Wed 15-Aug-2012 CC F326

Next Meeting

1:30 PM Wed 29-Aug-2012 CC F326

Minutes

Present: Elliott W, Simon T, Dave L, Hovanes E, Vanik K, Bryan M.

Mantis

Elliott urges everyone to actively use the Mantis system, as it has proven quite useful for the Offline effort and for him.

Computer System Manager

A while back Elliott talked with Eugene and Andy Kowalski about a computer system manager for Hall D. Originally we thought we'd need someone this fall, but now some time this spring or early summer seems more appropriate. Andy is happy with the idea of matrixing one of his staff to Hall D, ramping up the percentage from small to 100%, staying at 100% for maybe a year, then ramping down. Andy agreed that 100% means that the person effectively works for us, not the CC. We'll revisit this in the late fall. Hovanes pointed out that only a few people in the CC are qualified to take this on.

Possible tasks for system manager: specify and purchase computers, RAID system and file server, install and set up network equipment, set up computer monitoring system, write software to transfer data to silo, etc.

FDC test

Bryan doesn't know of any real progress on the FADC125 readout problem. They may be completely rewriting the FPGA software for the new 2ESST. No one knows how much time Gerard has for this. Beni is dead-in-the-water on this.

JInventory

Elliott is almost done with the procedures that manage the housing and other hierarchies. The hierarchies are managed in a separate table, so there will be no change to the base inventory table. Feel free to enter production information into the database.

JANA and Test Data

Dave is working on formalizing a way to read EVIO data into JANA and place it into standard objects that will be the starting point for event reconstruction. Following the JANA philosophy, the user of the data will not have to know anything about the EVIO file itself as the sole interface to the data will be via the objects. So far he can handle FADC250 and F1TDC data in most (all?) readout modes. He's been using FCAL and BCAL test data to exercise the code.

The question arose as to whether we want to standardize bank tag/num conventions for all test systems.

Counting House

Hovanes reported that work is nearly complete on wiring the Rack Room. When complete the racks will be installed and bolted down. At that point we will contract to have fibers run from the Telecom Room switch to the Rack Room and we'll install our switch.

DED

Andrew starts classes soon so he is finishing up. He is nearly done implementing two-way communication between DED and JANA.

RootSpy

We met with Sergey Boyarinov and came to agreement on sharing code. He is happy to use our variant of his original code as long as the original functionality is maintained. He plans to move histogram data using the ET system whereas we plan to use cMsg (out-of-band), but both can be accomodated. Rob is moving the hist data from local to shared memory so that a separate process from the ROC can serve the data out. Thus if something goes wrong the ROC shouldn't be affected.

Elog