Difference between revisions of "August 12, 2009 Software"

From GlueXWiki
Jump to: navigation, search
m (Text replacement - "/halldweb1.jlab.org/" to "/halldweb.jlab.org/")
 
(4 intermediate revisions by the same user not shown)
Line 39: Line 39:
 
===HDGeant crashes===
 
===HDGeant crashes===
  
We discussed whether the problem Beni has reported in the past is the same as that reported by Aram Teymurazyan??? this past week. This discussion will continue on the email list. Beni will however send out a patch for the original problem.
+
We discussed whether the problem Beni has reported in the past is the same as that reported by Aram Teymurazyan this past week. This discussion will continue on the email list. Beni will however send out a patch for the original problem.
  
 
===GlueX Software Task List===
 
===GlueX Software Task List===
  
The ???list has been ???updated to reflect the discussion from the last meeting.
+
The [http://www.jlab.org/Hall-D/software/Software_tasks.php list] has been [https://mailman.jlab.org/pipermail/halld-offline/2009-August/000009.html updated] to reflect the discussion from the last meeting.
  
==GPU title???==
+
==Scientific Computing and GPU's==
  
Jie Chen presented an overview of how the JLab LQCD computing group is trying to ???use GPU's??? to upgrade the processing power on its farm. They have tested hardware using the nVidia technology and have demonstrated good performance.
+
Jie Chen presented an overview of how the JLab LQCD computing group is trying to [https://halldweb.jlab.org/talks/2009-08/gpu-halld.pptx use GPU's] to upgrade the processing power on its farm. They have tested hardware using the nVidia technology and have demonstrated good performance.
  
 
==non-JLab Farms==
 
==non-JLab Farms==
Line 53: Line 53:
 
David reviewed the [[Computer Farms|wiki page]] he created after polling the collaboration about computing resources at their respective home institutions that might be available for GlueX work. He is thinking that we should try to gain some experience on one or two of these sites so that we can deploy those resources when needed. There was some discussion about how hard we should be working on this. Streamlining general access to these resources should clearly be studied, but it may not be the highest priority task at this time.
 
David reviewed the [[Computer Farms|wiki page]] he created after polling the collaboration about computing resources at their respective home institutions that might be available for GlueX work. He is thinking that we should try to gain some experience on one or two of these sites so that we can deploy those resources when needed. There was some discussion about how hard we should be working on this. Streamlining general access to these resources should clearly be studied, but it may not be the highest priority task at this time.
  
=disk discussion???=
+
==JLab Disk Resources for Hall D==
  
Mark reviewed the types and intended purposes of [[???|disk resources at JLab]]. This generated discussion on several topics:
+
Mark reviewed the types and intended purposes of [[Disk Space for Hall D at JLab|disk resources at JLab]]. This generated discussion on several topics:
  
 
* We need to document the procedure for mounting work disks on JLab desktop machines.
 
* We need to document the procedure for mounting work disks on JLab desktop machines.
Line 62: Line 62:
 
* We need to document a simple procedure for archiving files onto the tape library.
 
* We need to document a simple procedure for archiving files onto the tape library.
  
eugene brings up items from last time
+
==High-Priority Tasks==
  
dave will run poisson and do tracking analysis
+
We discussed the status of high priority projects that Eugene identified last time.
superfish.lanl.gov
+
 
 +
* Simon has done some work on studying an alternate geometry for the FDC. He will present his work soon.
 +
* Dave will run poisson and do tracking analysis on the resulting alternate magnetic fields.
 +
 
 +
==New Action Items==
 +
 
 +
* Patch the HDGeant problem with FDC index 0 -> Beni
 +
* All items discussed in the "JLab Disk Resources for Hall D" section above.
 +
* All items discussed in the "High-Priority Tasks" section above.
 +
 
 +
===maintaining the action item list===
 +
 
 +
We decided that each person should manage their own items on the [[Action_Items_From_Hall-D_Software_Meetings|action item list]], i. e., when you finish the action called for in the item, you also go to the wiki page and cross it off the list (actually, render it in overstrike font) yourself.

Latest revision as of 10:34, 31 March 2015

Location

The meeting will be at 2:00 pm in CEBAF Center F326/327.

Telephone

To connect by telephone:

  1. dial:
    • 800-377-8846 : US
    • 888-276-7715 : Canada
    • 302-709-8424 : International
  2. enter participant code: 77438230# (remember the "#")

Video Conferencing

A) ESNet: 8542553
B) EVO: direct meeting link

Agenda

  1. Review minutes from July 29, 2009 Software meeting
  2. Announcements
    • JLab email lists: Majordomo -> Mailman
    • Collaboration Meeting: Offline/Monte Carlo Session (45 min., 9/11)
  3. GPU use at JLab: Jie Chen
  4. non-JLab Farms: David
  5. Group disk quota increase and usage (work and cache for that matter): all
  6. Action item review
    • group review of list

Minutes

JLab: Jie Chen, Ying Chen, Eugene Chudakov, Mark Ito (chair), David Lawrence, Sascha Somov, Simon Taylor, Elliott Wolin, Beni Zihlmann
University of New Hampshire: Maurik Holtrop

Review minutes from last meeting

We went over minutes from the July 29, 2009 Software meeting

HDGeant crashes

We discussed whether the problem Beni has reported in the past is the same as that reported by Aram Teymurazyan this past week. This discussion will continue on the email list. Beni will however send out a patch for the original problem.

GlueX Software Task List

The list has been updated to reflect the discussion from the last meeting.

Scientific Computing and GPU's

Jie Chen presented an overview of how the JLab LQCD computing group is trying to use GPU's to upgrade the processing power on its farm. They have tested hardware using the nVidia technology and have demonstrated good performance.

non-JLab Farms

David reviewed the wiki page he created after polling the collaboration about computing resources at their respective home institutions that might be available for GlueX work. He is thinking that we should try to gain some experience on one or two of these sites so that we can deploy those resources when needed. There was some discussion about how hard we should be working on this. Streamlining general access to these resources should clearly be studied, but it may not be the highest priority task at this time.

JLab Disk Resources for Hall D

Mark reviewed the types and intended purposes of disk resources at JLab. This generated discussion on several topics:

  • We need to document the procedure for mounting work disks on JLab desktop machines.
  • We need a policy on how to archive mechanical and electronics documents.
  • We should survey other groups at the lab on their group disk usage.
  • We need to document a simple procedure for archiving files onto the tape library.

High-Priority Tasks

We discussed the status of high priority projects that Eugene identified last time.

  • Simon has done some work on studying an alternate geometry for the FDC. He will present his work soon.
  • Dave will run poisson and do tracking analysis on the resulting alternate magnetic fields.

New Action Items

  • Patch the HDGeant problem with FDC index 0 -> Beni
  • All items discussed in the "JLab Disk Resources for Hall D" section above.
  • All items discussed in the "High-Priority Tasks" section above.

maintaining the action item list

We decided that each person should manage their own items on the action item list, i. e., when you finish the action called for in the item, you also go to the wiki page and cross it off the list (actually, render it in overstrike font) yourself.