Difference between revisions of "2012 Software Review Planning Meeting: Mar. 2, 2012"

From GlueXWiki
Jump to: navigation, search
(Agenda)
Line 49: Line 49:
 
=Minutes=
 
=Minutes=
  
 +
Attendees: Eugene C., Mark I., David L., Curtis M.
  
Attendees: ``TBD``
+
== Status of Review Organization ==
 +
No change as to the status regarding charge or reviewers. David and Eugene both forwarded names of potential reviewers
 +
to Bob McKeown. Bob acknowledged receipt in both cases, but no more.
 +
 
 +
== Cross-Hall Collaboration ==
 +
A draft report from the internal review of the CLAS12 software effort has come out. David was on the review panel and thus,
 +
contributed some to the report. There were serious concerns expressed in the report that CLAS12 would not do well in the
 +
review unless more effort was directed towards the software than is currently being done. They are now taking steps to
 +
address this.
 +
 
 +
Rolf has tentatively scheduled a follow up meeting to his late January "brainstorming" session to identify areas of cooperation
 +
between the halls on software. Halls were tasked to meet and discuss certain topics so we could come back at a later meeting
 +
and report results. Meetings took place between Halls B and D on the topics assigned specifically for them to discuss (Tracking
 +
and JANA/CLARA). No meeting has occurred to discuss the topics assigned to all 4 halls. Two of these topics were to have
 +
meetings organized by the halls themselves while the other 3 were to be organized by Graham Heyes. David volunteered to
 +
arrange a meeting on the two non-Graham-assigned topics: Histogram monitoring and Single Event Displays.
 +
 
 +
Curtis mentioned that we could discuss our nightly cron jobs as part of the histogram monitoring discussion.
 +
 
 +
There was brief discussion on a common event format. In particular for the post-processed DST data. What Hall-D should do
 +
was discussed at the last offline meeting. The decision was to tentatively adopt the format UConn has developed and
 +
implemented in HDDM. Some work will need to be done to write out events in this format so estimates of DST disk space
 +
requirements can be made. We assume billions of processed events will need to be stored. (Some discussion of adopting
 +
the EventStore product from CLEO was also had at the Offline meeting.) It is also guesstimated at this
 +
point that the micro-DST size may be small enough that most universities will be able to host disk space to hold it.
 +
Also, one or two good GPU-based machines may allow for local analysis using PWA. The requirements will need to be
 +
worked out to back this up by the time of the review.
 +
 
 +
== Hall-D preparations for Review ==
 +
Curtis suggested that about 10 minutes of the time allotted to Hall-D be used at the beginning for an outside collaborator
 +
(most likely Curtis or Matt) to give an introduction to GlueX/Hall-D. At the same time, they and provide an outside researchers
 +
perspective on what is needed to get to the point of physics analysis. He will begin to work on some slides for this.
 +
 
 +
We also discussed the issue of Calibration/Alignment. This represents a large fraction of our estimated software effort (16%).
 +
We will need to focus on this some between now and the time of the review in order to present a clear plan. Curtis will
 +
begin adding it as an item for the individual detector reports at the bi-weekly meetings. The thought is that many of the requirements
 +
will originate from the detector groups who will be best positioned to develop them. A special session on this will also
 +
be added to the May collaboration meeting agenda. Curtis noted that Naomi will be able to spend time working on this
 +
for the CDC. Eugene noted that one of the first things we'll need to do is look up details of the cosmic spectrum so as to
 +
make reasonable estimates on rates etc. in the hall.
  
  
 
=Action Items=
 
=Action Items=
 +
 +
* Organize meeting with 4 halls to discuss Histogram Monitoring and Single Event Displays (David)
 +
* Start working on slides for introduction to Hall-D presentation at Software Review (Curtis)
 +
* Add Calibration/Alignment to standard bi-weekly meeting agenda (Curtis)

Revision as of 10:46, 9 March 2012

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

  • Previous Meeting (Feb. 24, 2012)
  • Status of Review Organization (JLab organizers)
  • Cross-Hall Collaboration
    • CLAS12 Review
      • Draft of report circulated
      • "Software Review Preparation" meeting today at 3pm in A110 (ReadyTalk participant code 7022780#)
    • Physics division effort to identify and promote cross-hall collaboration
Dear all,

    We plan a followup software review preparation meeting
tentatively on Wednesday March 14th, from 3:30 to 5:00 pm (EST).
The hope is to then go over the "less-than-one-summarizing-page"
to stare at on the sub-meetings that many of you have been doing.
We likely will also discuss a bit more preparations and dry runs
for the final software review.

    For now, the meeting is tentatively plannned for A110,
but this may change.

                    Best regards,    Rolf 

Minutes

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

Status of Review Organization

No change as to the status regarding charge or reviewers. David and Eugene both forwarded names of potential reviewers to Bob McKeown. Bob acknowledged receipt in both cases, but no more.

Cross-Hall Collaboration

A draft report from the internal review of the CLAS12 software effort has come out. David was on the review panel and thus, contributed some to the report. There were serious concerns expressed in the report that CLAS12 would not do well in the review unless more effort was directed towards the software than is currently being done. They are now taking steps to address this.

Rolf has tentatively scheduled a follow up meeting to his late January "brainstorming" session to identify areas of cooperation between the halls on software. Halls were tasked to meet and discuss certain topics so we could come back at a later meeting and report results. Meetings took place between Halls B and D on the topics assigned specifically for them to discuss (Tracking and JANA/CLARA). No meeting has occurred to discuss the topics assigned to all 4 halls. Two of these topics were to have meetings organized by the halls themselves while the other 3 were to be organized by Graham Heyes. David volunteered to arrange a meeting on the two non-Graham-assigned topics: Histogram monitoring and Single Event Displays.

Curtis mentioned that we could discuss our nightly cron jobs as part of the histogram monitoring discussion.

There was brief discussion on a common event format. In particular for the post-processed DST data. What Hall-D should do was discussed at the last offline meeting. The decision was to tentatively adopt the format UConn has developed and implemented in HDDM. Some work will need to be done to write out events in this format so estimates of DST disk space requirements can be made. We assume billions of processed events will need to be stored. (Some discussion of adopting the EventStore product from CLEO was also had at the Offline meeting.) It is also guesstimated at this point that the micro-DST size may be small enough that most universities will be able to host disk space to hold it. Also, one or two good GPU-based machines may allow for local analysis using PWA. The requirements will need to be worked out to back this up by the time of the review.

Hall-D preparations for Review

Curtis suggested that about 10 minutes of the time allotted to Hall-D be used at the beginning for an outside collaborator (most likely Curtis or Matt) to give an introduction to GlueX/Hall-D. At the same time, they and provide an outside researchers perspective on what is needed to get to the point of physics analysis. He will begin to work on some slides for this.

We also discussed the issue of Calibration/Alignment. This represents a large fraction of our estimated software effort (16%). We will need to focus on this some between now and the time of the review in order to present a clear plan. Curtis will begin adding it as an item for the individual detector reports at the bi-weekly meetings. The thought is that many of the requirements will originate from the detector groups who will be best positioned to develop them. A special session on this will also be added to the May collaboration meeting agenda. Curtis noted that Naomi will be able to spend time working on this for the CDC. Eugene noted that one of the first things we'll need to do is look up details of the cosmic spectrum so as to make reasonable estimates on rates etc. in the hall.


Action Items

  • Organize meeting with 4 halls to discuss Histogram Monitoring and Single Event Displays (David)
  • Start working on slides for introduction to Hall-D presentation at Software Review (Curtis)
  • Add Calibration/Alignment to standard bi-weekly meeting agenda (Curtis)