Difference between revisions of "Software and Computing Review 5"

From GlueXWiki
Jump to: navigation, search
(Agenda)
m (Report from the Review Committee)
 
(10 intermediate revisions by 3 users not shown)
Line 1: Line 1:
 
__TOC__
 
__TOC__
  
=== Organizing Note from Rolf Ent ===
+
= November 27-28, 2018 at Jefferson Lab =
 +
 
 +
[https://www.jlab.org/indico/event/294/ Indico Site]
 +
== Report from the Review Committee ==
 +
[https://halldweb.jlab.org/doc-private/DocDB/ShowDocument?docid=3936 Report] archived as GlueX-Doc-3936 by C.A. Meyer, actual authors are David Brown (LBNL), Frank Golf (UNL), Adam Lyon (FNAL), and Arjun Shankar (ORNL).
 +
 
 +
== Earlier Reviews ==
 +
* [[Software Review 4]] JLab, November 10-11, 2016
 +
** [https://halldweb.jlab.org/doc-private/DocDB/ShowDocument?docid=3189 Final Report] for the 2016 Review
 +
* [[Software Review 3| February 2015 Software Review]]
 +
** [https://halldweb.jlab.org/doc-private/DocDB/ShowDocument?docid=2701 Final Report] for the 2015 Review
 +
* [[2013 Software Review|November 2013 Software Review]]
 +
** [https://halldweb.jlab.org/doc-private/DocDB/ShowDocument?docid=2382 Final Report] for the 2013 Review
 +
* [[2012 Software Review|June 2012 Software Review]]
 +
** [[Media:JLab12GeVSoftwareComputingReviewFinal20120912.pdf|Final Report from the 2012 Review]](PDF)
 +
 
 +
=Preparations=
 +
== All-Hall Meetings ==
 +
*[[Software Review 5 - All Hall - November 6 | November 6, 2018]]
 +
 
 +
== Presentations ==
 +
* Part A: GlueX/Hall-D: Analysis and Physics [https://halldweb.jlab.org/doc-private/DocDB/ShowDocument?docid=3813 GlueX-Doc 3813] Curtis Meyer
 +
 
 +
== Organizing Note from Rolf Ent ==
  
 
<font color="blue">
 
<font color="blue">
Line 34: Line 57:
 
** 9:30 – 10:00 SBS Software and Computing - Seamus Riordan (ANL)? (include *** status of GEM tracking/reconstruction software & validation)
 
** 9:30 – 10:00 SBS Software and Computing - Seamus Riordan (ANL)? (include *** status of GEM tracking/reconstruction software & validation)
 
* 10:00 – 10:15 Break
 
* 10:00 – 10:15 Break
** 10:15 – 10:45 GlueX/Hall D Overview and Progress: A Blueprint for Physics - Curtis Meyer / David Lawrence
+
** <font color="brown">10:15 – 10:45 GlueX/Hall D Overview and Progress: A Blueprint for Physics - Curtis Meyer / David Lawrence </font>
*** (include what the detectors are, the schedule for the next five years, computing/software experience so far, lessons learned, prospects for high-intensity run, tracking versus recommendations/milestones, efforts and planning for off-site computing, estimates of on-site Computing and Storage Requirements for 5 years)
+
*** <font color="brown"> (include what the detectors are, the schedule for the next five years, computing/software experience so far, lessons learned, prospects for high-intensity run, tracking versus recommendations/milestones, efforts and planning for off-site computing, estimates of on-site Computing and Storage Requirements for 5 years)</font>
 
** 10:45 – 11:15 CLAS12/Hall B Overview and Progress - Stepan Stepanyan?
 
** 10:45 – 11:15 CLAS12/Hall B Overview and Progress - Stepan Stepanyan?
 
*** (include what the detectors are, the schedule for the next five years, software experience from first 12-GeV experiments, path to publication, high-level overview and tracking versus recommendations/milestones, efforts and planning for off-site computing, estimates of on-site Computing and Storage Requirements for 5 years)
 
*** (include what the detectors are, the schedule for the next five years, software experience from first 12-GeV experiments, path to publication, high-level overview and tracking versus recommendations/milestones, efforts and planning for off-site computing, estimates of on-site Computing and Storage Requirements for 5 years)
Line 42: Line 65:
 
** 12:15 – 12:45 Experimental Computing Overview - Graham Heyes
 
** 12:15 – 12:45 Experimental Computing Overview - Graham Heyes
 
***(roll-up of requirements and facility planning, roll-up of off-site planning assumptions, DOE/NP report)
 
***(roll-up of requirements and facility planning, roll-up of off-site planning assumptions, DOE/NP report)
* 12:45 – 2:00 Working Lunch (Executive session)
+
* 12:45 – 2:00 Working Lunch (Executive session)
 +
** 2:00 – 2:30 Scientific Computing and Scope at Jefferson Lab - Chip Watson
 +
*** (overview of current resources, and the process by which we arrived at this point including optimizing for science (compute, storage and bandwidth, memory, networking), ongoing evolution of the computing environment to support swings in demand and to integrate available ("free") external resources, near-term (FY2019) evolution (cloud, increasing offsite resources), including expected changes in computer architecture (component) and how we choose likely future winners)
 +
** 2:30 – 3:00 Scientific Computing Implementation - Sandy Philpott
 +
*** (Current capacity (compared to requirements) and FY2019 likely purchases to respond to requirements, Operations (everything that serves the users: queues, allocations, fair share, trouble tickets), overview of work flow tools: Auger, Jasmine, SWIF (present and FY2019 evolution), management (decision making) and software support (Physics div. "volunteers"))
 +
** 3:00 – 4:30 Outlook and Gaps for Future Nuclear Physics Program - (3 talks, 30 minutes each, <50% presentation)
 +
*** 1) Gaps for Possible Future Experimental Program Steve Wood?
 +
(include GEANT4 needs, TDIS/streaming, MOLLER, SoLID, etc.)
 +
*** 2) Gaps for Possible Future Scientific Computing Chip Watson
 +
(include planning for 24/7 computing support, plans/hurdles for advanced scientific computing in out-years)
 +
*** 3) Gaps for Possible Future Phenomenology Nobuo Sato/Jianwei Qiu?
 +
(include state-of-the-art global fit/phenomenology, further software
 +
development needed for radiative corrections & 3D global fits, needs)
 +
 
 +
* 4:30 – 4:45 Break
 +
**  4:45 – 6:30 Executive session
 +
**  6:30 – 6:45 Questions
 +
*  7:00 – 9:00 Reception
 +
 
 +
* Wednesday, November 28
 +
** 8:00 – 10:00 Q&A
 +
** 10:00 – 12:00 Executive session – prepare draft
 +
** 1:00 –  1:30 Close Out

Latest revision as of 12:56, 7 March 2019

November 27-28, 2018 at Jefferson Lab

Indico Site

Report from the Review Committee

Report archived as GlueX-Doc-3936 by C.A. Meyer, actual authors are David Brown (LBNL), Frank Golf (UNL), Adam Lyon (FNAL), and Arjun Shankar (ORNL).

Earlier Reviews

Preparations

All-Hall Meetings

Presentations

  • Part A: GlueX/Hall-D: Analysis and Physics GlueX-Doc 3813 Curtis Meyer

Organizing Note from Rolf Ent

Dear all,

as you likely have heard from your Hall Leaders, we plan another version of the software and computing reviews we have had preparing for the 12 GeV era. Of course, the scenery changed as we are now in official 4-Hall operations, so you will see the emphasis of the review shift more to the experimental computing status and outlook for the next five years. However, we do plan to pick up some remnants of software preparation and status of items that were not completely ready yet at the time of the last review - these are related to SBS and CLAS12.

Please see attached the charge and the suggested agenda. Speaker names are of course tentative and we need the Hall input, especially for those with a "?". As before, I added some commentary that is hopefully useful on what to include/expect in the various presentations.

As last time we will have a generic talk on Halls A/C, Hall B and Hall D status with updates etc. And then as said there are "deep dive" talks included for SBS and for Hall B/CLAS12. Graham will have a presentation on the 'roll-up" computing requirements, which also include plans for off site Monte Carlo. The Hall software representatives will need to work with Graham to provide updates on the projections. All talks are plenary as that was the preference of the reviewers last time.

In the afternoon the emphasis is more on the scientific computing, followed by a session more forward looking to potential gaps appearing on the horizon.

Given the change of emphasis of the review this time, we will have some new reviewers. Torre Wenaus will as before be the chair, and we plan to at least have one repeat reviewer for consistency, but we also plan to have one reviewer each with expertise in applications and computing facilities.

I would like the usual Hall and user software suspects to look at the agenda, and come back to me with any change of speaker suggestions. My guess is beyond that stage we plan one initial "dry run" just to see/learn how different groups "fill in" their talks, and then as before will load draft presentations and comment/improve/iterate while the draft talks come in.

Best regards, Rolf

Agenda

Jefferson Lab 12-GeV Experimental Computing Review November 27-28, 2018 L102/104 CEBAF Center

  • Tuesday, November 27
    • 8:00 – 8:45 Executive Session – also includes few slides on where we are with 12-GeV
    • 8:45 – 9:00 Introduction – General JLab Computing and Experimental Software Structure - Amber Boehnlein
    • 9:00 – 9:30 Halls A and C Overview and Progress - Ole Hansen / Mark Jones
      • (include what the detectors are, the schedule for the next five years, software experience from first 12 GeV experiments, path to publication, high-level overview and tracking versus recommendations/milestones, estimates of Computing and Storage Requirements for 5 years)
    • 9:30 – 10:00 SBS Software and Computing - Seamus Riordan (ANL)? (include *** status of GEM tracking/reconstruction software & validation)
  • 10:00 – 10:15 Break
    • 10:15 – 10:45 GlueX/Hall D Overview and Progress: A Blueprint for Physics - Curtis Meyer / David Lawrence
      • (include what the detectors are, the schedule for the next five years, computing/software experience so far, lessons learned, prospects for high-intensity run, tracking versus recommendations/milestones, efforts and planning for off-site computing, estimates of on-site Computing and Storage Requirements for 5 years)
    • 10:45 – 11:15 CLAS12/Hall B Overview and Progress - Stepan Stepanyan?
      • (include what the detectors are, the schedule for the next five years, software experience from first 12-GeV experiments, path to publication, high-level overview and tracking versus recommendations/milestones, efforts and planning for off-site computing, estimates of on-site Computing and Storage Requirements for 5 years)
    • 11:15 – 12:15 CLAS12 Software and Computing Deep Dive - Raffaella De Vita / G. Heyes
      • (maybe split in 2 talks? Talk about CLARA software status, gaps, items to complete, changes made in computing requirements, software, organization since last review, recent successes, etc.)
    • 12:15 – 12:45 Experimental Computing Overview - Graham Heyes
      • (roll-up of requirements and facility planning, roll-up of off-site planning assumptions, DOE/NP report)
  • 12:45 – 2:00 Working Lunch (Executive session)
    • 2:00 – 2:30 Scientific Computing and Scope at Jefferson Lab - Chip Watson
      • (overview of current resources, and the process by which we arrived at this point including optimizing for science (compute, storage and bandwidth, memory, networking), ongoing evolution of the computing environment to support swings in demand and to integrate available ("free") external resources, near-term (FY2019) evolution (cloud, increasing offsite resources), including expected changes in computer architecture (component) and how we choose likely future winners)
    • 2:30 – 3:00 Scientific Computing Implementation - Sandy Philpott
      • (Current capacity (compared to requirements) and FY2019 likely purchases to respond to requirements, Operations (everything that serves the users: queues, allocations, fair share, trouble tickets), overview of work flow tools: Auger, Jasmine, SWIF (present and FY2019 evolution), management (decision making) and software support (Physics div. "volunteers"))
    • 3:00 – 4:30 Outlook and Gaps for Future Nuclear Physics Program - (3 talks, 30 minutes each, <50% presentation)
      • 1) Gaps for Possible Future Experimental Program Steve Wood?

(include GEANT4 needs, TDIS/streaming, MOLLER, SoLID, etc.)

      • 2) Gaps for Possible Future Scientific Computing Chip Watson

(include planning for 24/7 computing support, plans/hurdles for advanced scientific computing in out-years)

      • 3) Gaps for Possible Future Phenomenology Nobuo Sato/Jianwei Qiu?

(include state-of-the-art global fit/phenomenology, further software development needed for radiative corrections & 3D global fits, needs)

  • 4:30 – 4:45 Break
    • 4:45 – 6:30 Executive session
    • 6:30 – 6:45 Questions
  • 7:00 – 9:00 Reception
  • Wednesday, November 28
    • 8:00 – 10:00 Q&A
    • 10:00 – 12:00 Executive session – prepare draft
    • 1:00 – 1:30 Close Out