(back to the list of all DMLT meeting minutes)

NOTE: For this meeting and future dates we are conducting the meeting on Blue Jeans (use web browser or below):

Web Browser:

https://bluejeans.com/293724745/

> Connecting directly from a room system?

> 1) Dial: 199.48.152.152 or bjn.vc
> 2) Enter Meeting ID: 293724745 -or- use the pairing code

 > 
> Just want to dial in? (all numbers)
> 1) Direct-dial with my iPhone or 
> +1 408 740 7256+1 408 740 7256 
> +1 888 240 2560+1 888 240 2560 (US Toll Free)
> +1 408 317 9253+1 408 317 9253 (Alternate Number)
> 
> 2) Enter Meeting ID: 293724745

Date

Attendees

(sorry if I missed anyone)

 

Goals
  • Weekly coordination meeting to discuss and resolve management items, project communications, external contacts, etc.

Discussion items

 

Item
Who
Notes
LSST 2015 MeetingKian-Tat Lim
DM contributions have been added to LSST 2015 agenda page. Not clear if some sessions others requested (Calibration, Camera-related) have been added.
KPM session is working session on adding code to pipelines to produce measurements.
Two decision-making sessions (python and policies) are in agenda.
Planning session is not scheduled, will do as much as possible before-hand, but still need session.
Donald Petravick reports that Athol Kembell and new NCSA Systems Management Lead will attend.
OCS Working Session DebriefKian-Tat Lim

Attended working session at NCSA last week. IInterfaces focused, planned integration tests to demonstrate interfaces are working. Resolved DMCS components that will interact with OCS, discussed configuration, have reasonable and suitable agreement that will work for DM. Took current OCS middleware based on OpenSplice DDS, got proof-of-concept driver programs to transmit (commands and telemetry) between OCS - CCS, and OCS - DMCS simulator (from Java to C++ to python). CCS is ahead of schedule on simulator, takes commands, sends telemetry in approximately right timing, DM to do similar by end of 2015. Capturing agreements and getting through Change Control is underway, due September. DM used boost python based wrapping, also used SWIG wrapper (still has some problems). Robert Lupton says we need to demonstrate locking, other mechanisms for calibration purposes.

LDM-240 updates

Jeff Kantor

Kian-Tat Lim

Work in progress on AMCL request for preliminary look at the material by  . Jeff Kantor, Simon Krughoff produced LDM-240 style tables from scripts created by Frossie Economou, Tim Jenness. Scripts now generate text, html-linked, whites space padded or csv tables of milestones. Jeff Kantor produced Excel file from csv (attached to this page). There is still some sorting by cycle and merging by WBS needed. Considering confluence page equivalent of Excel file. Next version to add Key Peformance Metrics tables.

Kian-Tat Lim, Mario Juric need to review and define issues to address and discuss prior to Bremerton.

Jim Bosch  has created tool that generates dependency graph of blocking relationships in DLP data. Works with any JIRA query, works well with a single WBS.

FY16 budgets and staffing assumptions

Jeff Kantor

Jeff Kantor and Kevin Long prepared material for discussion with T/CAMs, verifyied assumptions about rest of FY15. Several institutions have large (100%+) jumps in budget between FY15 and FY16, not clear if hiring, spending can proceed at that pace. Will send out "upper limit", ask T/CAMs to provide actual planned budget which will probably be less, contracts will be amended to "realistic" number. This is linked to verifying resource loading implied by LDM-240.

Meeting schedulesJeff Kantor
LSST 2016 in Tucson, week of August 15
Project Team 2016 meeting in Bay Area week of February 15
  • Also recall DMLT in Princeton  -  

PST debrief

Kian-Tat Lim

Mario Juric

Steven Kahn reported there will be a proposal for EUCLID/WFIRST/LSST Joint Processing, with impacts on project and DM. Scientifically obvious to combine image-level processing. Complicated since separate agencies, separate projects, separate continents. Group being assembled to discuss in 4 phases:

1: Concept: What, who is necessary? (telecons)

2: Define algorithms, data movement, processing changes

3: Build out systems, additions to existing systems

4: Operate this way

Summer 2015 progress/closeoutKian-Tat LimOur planned conclusion date is  . Need to discuss what will be ready by LSST 2015 meeting, closeout. Preparation for Winter 2016. Merge HSC code back in and KPM measurements are key goals. Lots of work will be done prior to LSST 2015, and will try to finish off at meeting. Not planning for significant demonstrations of the release. We have some sessions in which status will be presented.
   

 Action items