July 10, 2006 Software

From GlueXWiki
Jump to: navigation, search

This was a Software meeting with one topic: to discuss the items on the software task list.

Below ae the minutes which can also be found at http://www.jlab.org/~davidl/HallD/Minutes/ .

In attendance:

David L., Elliott W. Simon T., Matt S., Richard J., Nikolay, K., Zisis P., Rafael H., Blake L.

The meeting agenda is posted on the Wiki at:



1 Review Software Task list 2 AOT

It was suggested that we add a discussion of Software Coding Conventions to the agenda. In the end though we ran over time and decided to delay that until the next meeting.

Future meeting suggestions: - Smart Pointers - BCAL beamtest simulations


Review Software Task list

The software task list can be found at: http://www.jlab.org/Hall-D/software/Software_tasks.php

The entire software task list was reviewed. Some highlights from the discussions follow:

HDGeant: A new document was recently added to the DocDB summarizing the current state of the detector geometry as defined to HDGeant. All systems except possibly the TOF and UPV need to be reviewed and updated by the persons responsible for each subsystem. Richard noted that he knows of at least a few differences between the geometry currently in HDGeant and the baseline design. He also noted that the current geometry is almost certainly an underestimate of the amount of material in the detector. This will be important to get realistic background rates from HDGeant.

HDDM: Richard noted that work is underway on a new C++, stream oriented wrapper for HDDM that includes features such as compression/decompression.

HDDS: Essentially stable now. It now has the ability to turn off magnetic field in the BCAL (for example) to speed up processing time. There was some discussion about how CLAS12 is still strongly considering adopting it. They are looking at the possibility of specifying hits read-out in XML as well so one does not need to write accompanying code with changes to the geometry. We agreed that we will wait to see what they come up with and then decide if it will be useful for us.

Web Portal: Zisis reported that they are starting to work on upgrading the portal from a patched version 5 to version 7. Richard asked about certificate support and Zisis said it did have one standard in v5, but isn't sure what v7 offers.

GRID: Richard has now installed his OpenShop package on clusters at Regina, Indiana, FSU and UConn. It can be used by remote sites to submit simulation jobs. OpenShop keeps a record of the jobs run and allows access to them from other users. Interested persons should contact Richard about getting a certificate and learning how to use it. Matt S. reminded us about the e-mail sent out a few weeks ago about the grant program "Physics in the Information Frontiers". He suggested we should put a proposal together for developing software for doing distributed simulations across remote clusters. The proposal would involve maily manpower.

Calibration Database: Nikolay reported that he has been working on this and will present what he's done in 2 weeks.

BCAL reconstruction: Work is stalled on this in order to focus on the beam test.

BCAL Simulations: The simulation of the BCAL beam test continues to be refined with things like the CLAS target being added. Blake L. is doing several simulations with the various comfigurations of the BCAL module during the test. He will present something at the next meeting (in two weeks).

Cherenkov: This is stalled completely. Richard noted that he should get at least a 3 month "heads up" to incorporate a Cherenkov detector into the simulation before anyone should expect to use it.

FCAL: Matt S. reported that he is working to finalize on a channel count. He and Richard will communicate offline to discuss the possibility of porting Richard's Rad-Phi C++ reconstruction code for the FCAL.

FDC: Work on the reconstruction software has stopped with the departure of Craig B. Simon has done some research on reconstruction techniques and is well positioned to serve in an advisory role if a person can be found to write the code.

TOF: Matt S. reported that they still have some cosmic tests to do. Indiana plans to try and locate a graduate student to work on the code under Ryan Mitchell's guidance.

Trigger Simulation: This is now considered an orphan project. David L. agreed to discuss this with Dave Doughty the next time he saw him.

--- Suggested changes to Software task list page ---

Background simulations: Richard has recently done some simulations of the backgrounds in the tagger building due to the electron beam dump. The background rates in the GlueX detector need to be studied with the more realistic magnetic field map and a geometry that contains more materials. This is one of our top priorities.

Add Elog item with Greg Riccardi as contact.