Difference between revisions of "2012 Software Review Planning Meeting: Apr. 27, 2012"

From GlueXWiki
Jump to: navigation, search
(Agenda)
m (Text replacement - "http://argus.phys.uregina.ca/cgi-bin/private" to "https://halldweb.jlab.org/doc-private")
 
(2 intermediate revisions by one other user not shown)
Line 34: Line 34:
 
#*:
 
#*:
 
# Organization of Hall-D preparations for Review
 
# Organization of Hall-D preparations for Review
#* Introductory talk ([http://argus.phys.uregina.ca/cgi-bin/private/DocDB/ShowDocument?docid=1949 preliminary sildes]) with notes.
+
#* Introductory talk ([https://halldweb.jlab.org/doc-private/DocDB/ShowDocument?docid=1949 preliminary sildes]) with notes.
 
#* Numbers for the table
 
#* Numbers for the table
 
#** Software Manpower within collaboration ([https://halldsvn.jlab.org/repos/trunk/docs/offline/ProjectProgress/manpower_survey.pdf PDF],[https://halldsvn.jlab.org/repos/trunk/docs/offline/ProjectProgress/manpower_survey.ods ODS])
 
#** Software Manpower within collaboration ([https://halldsvn.jlab.org/repos/trunk/docs/offline/ProjectProgress/manpower_survey.pdf PDF],[https://halldsvn.jlab.org/repos/trunk/docs/offline/ProjectProgress/manpower_survey.ods ODS])
Line 40: Line 40:
 
#** Computing Resources within the Collaboration ([https://halldsvn.jlab.org/repos/trunk/docs/offline/ProjectProgress/ComputingResources.pdf PDF] , [https://halldsvn.jlab.org/repos/trunk/docs/offline/ProjectProgress/ComputingResources.ods ODS])
 
#** Computing Resources within the Collaboration ([https://halldsvn.jlab.org/repos/trunk/docs/offline/ProjectProgress/ComputingResources.pdf PDF] , [https://halldsvn.jlab.org/repos/trunk/docs/offline/ProjectProgress/ComputingResources.ods ODS])
 
#** Simulation requirements (GRID, CPU, disk space)
 
#** Simulation requirements (GRID, CPU, disk space)
 +
#***  [https://halldsvn.jlab.org/repos/trunk/home/marki/gluex/docs/comp_mod/computing_model.xls Spreadsheet] ([[Dave's preliminary comment's on offline computing spreadsheet numbers|comments]])
 
#** Reconstruction requirements (tape, CPU, disk)
 
#** Reconstruction requirements (tape, CPU, disk)
 
#** Analysis requirements (offsite transfer bandwidth/volume, CPU/GPU)
 
#** Analysis requirements (offsite transfer bandwidth/volume, CPU/GPU)
Line 50: Line 51:
 
=Minutes=
 
=Minutes=
  
''TBD''
+
Attendees: David L., Mark I., Matt S., Eugene C., Curtis M.
 +
 
 +
== Charge ==
 +
 
 +
* A [[Media:20120426_JLab_review_Charge_v3.pdf|charge]] has been posted. Only minor tweaks are expected.
 +
* It was unclear what item 2c: "''Are there adequate plans for transitioning from a development phase into a deployment and operations phase?''" has to do with management
 +
* Charge mentions contingency (item 2b) but we don't address that directly in our slide outlines. This should be added
 +
* Workshops should be highlighted to address item 1e: "''Have the collaborations identified effective and appropriate mechanisms to support utilization of the software by the entire collaboration ?''"
 +
* Slide outlines should be updated prior to May 1st meeting to make sure they address all elements of the charge. Curtis and Dave will do this.
 +
 
 +
== Computing Numbers ==
 +
There was very lengthy discussion over Mark's spreadsheet that calculates the required resources for steady state GlueX running. Some notes from that discussion:
 +
* The running period covered may exceed currently approved running days, but that's OK
 +
* Some analysis items may not be easily identifiable in the spreadsheet. In other words, if one wrote down the final stages of the Physics Analysis plan, it might not be easy to see where the resources required for that are covered in the spreadsheet.
 +
** Curtis suggested another, smaller spreadsheet for this
 +
** Matt will attempt to create such a spreadsheet
 +
* In the spreadsheet lingo, stream=skim
 +
* Eugene noted that the numbers will be different for low luminosity running and high luminosity running. He suggested they be broken out into 1st year, 2nd year, 3rd year running
 +
** Curtis suggested using phase I, II, and III running
 +
*** Phase I engineering run (1 PAC month)
 +
*** Phase II 1 PAC month
 +
*** Phase III 2 PAC months
 +
*** Phase IV High Intensity running with L3 trigger (internal only)
 +
** We should request resources for phase II up front (low luminosity)
 +
** Phase II may be "thought and manpower limited"
 +
* There was some discussion about the time requirements for a real calibration/reconstruction/analysis pass. Namely, delays due to "thinking time" or when computers were idle waiting for human work to complete. Also, how long are we practically wanting to wait for a calibration pass to complete
 +
** Mark said a resource timeline can be developed if a time-sensitive model is made. It complicates things quite a bit since the model will add lots of parameters
 +
 
 +
== Milestones==
 +
* Change "1 day low luminosity" to "1 day 10^7 gamma/s"
 +
* Need more quantitative values to be attached to milestones. Some milestones may just include measurement of these quantities
 +
** Efficiency
 +
** Resolution
 +
** False tracks or false clusters
 +
* We should capture the work we're doing in table
 +
** include already passed milestones
 +
** add milestones from activity schedule
 +
*** find things we think we can achieve
 +
***# % false tracks in multi-track events
 +
***# % efficiency in multi-track events
 +
 
 
=Action Items=
 
=Action Items=
  
''TBD''
+
# Update slides outlines to address all items in charge (Curtis, David)
 +
# Create spreadsheet with requirements for Physics Analysis (Matt)
 +
# Update milestones list

Latest revision as of 16:58, 24 February 2017

Main Hall-D Software Review Page


Connecting

Phone:

  • +1-866-740-1260 : US+Canada
  • +1-303-248-0285 : International
  • then enter participant code: 3421244# (remember the "#").
  • or www.readytalk.com (and code without the #)

Agenda

  1. Previous Meeting (Apr. 20, 2012)
  2. Status of Review Organization (JLab organizers)
    • Committee (Wenaus(chair), Marchand, McBride, Purske, +1??)
    • Charge
    • Agenda
  3. Time-line for preparations
  4. Organization of Hall-D preparations for Review

Minutes

Attendees: David L., Mark I., Matt S., Eugene C., Curtis M.

Charge

  • A charge has been posted. Only minor tweaks are expected.
  • It was unclear what item 2c: "Are there adequate plans for transitioning from a development phase into a deployment and operations phase?" has to do with management
  • Charge mentions contingency (item 2b) but we don't address that directly in our slide outlines. This should be added
  • Workshops should be highlighted to address item 1e: "Have the collaborations identified effective and appropriate mechanisms to support utilization of the software by the entire collaboration ?"
  • Slide outlines should be updated prior to May 1st meeting to make sure they address all elements of the charge. Curtis and Dave will do this.

Computing Numbers

There was very lengthy discussion over Mark's spreadsheet that calculates the required resources for steady state GlueX running. Some notes from that discussion:

  • The running period covered may exceed currently approved running days, but that's OK
  • Some analysis items may not be easily identifiable in the spreadsheet. In other words, if one wrote down the final stages of the Physics Analysis plan, it might not be easy to see where the resources required for that are covered in the spreadsheet.
    • Curtis suggested another, smaller spreadsheet for this
    • Matt will attempt to create such a spreadsheet
  • In the spreadsheet lingo, stream=skim
  • Eugene noted that the numbers will be different for low luminosity running and high luminosity running. He suggested they be broken out into 1st year, 2nd year, 3rd year running
    • Curtis suggested using phase I, II, and III running
      • Phase I engineering run (1 PAC month)
      • Phase II 1 PAC month
      • Phase III 2 PAC months
      • Phase IV High Intensity running with L3 trigger (internal only)
    • We should request resources for phase II up front (low luminosity)
    • Phase II may be "thought and manpower limited"
  • There was some discussion about the time requirements for a real calibration/reconstruction/analysis pass. Namely, delays due to "thinking time" or when computers were idle waiting for human work to complete. Also, how long are we practically wanting to wait for a calibration pass to complete
    • Mark said a resource timeline can be developed if a time-sensitive model is made. It complicates things quite a bit since the model will add lots of parameters

Milestones

  • Change "1 day low luminosity" to "1 day 10^7 gamma/s"
  • Need more quantitative values to be attached to milestones. Some milestones may just include measurement of these quantities
    • Efficiency
    • Resolution
    • False tracks or false clusters
  • We should capture the work we're doing in table
    • include already passed milestones
    • add milestones from activity schedule
      • find things we think we can achieve
        1.  % false tracks in multi-track events
        2.  % efficiency in multi-track events

Action Items

  1. Update slides outlines to address all items in charge (Curtis, David)
  2. Create spreadsheet with requirements for Physics Analysis (Matt)
  3. Update milestones list