Date
at 9:30am PST
Tucson Location: LSST Conference Room
Videocom: IP 140.252.24.8 (Direct Dial)
Call 1-866-330-1200, Participant Number 518-2287#
Attendees
George, Kem, Cathy, Peter, Michael, Steve, Lynne, Andy
Agenda & Notes
Behavior of RestartCompleteSequences = True and what code actually does (Kem & Francisco).
- restartCompleteSequences - restarts ENTIRE sequence so in WFD it needs to finish all filters for a field before observing more in any filter.
- overflow - fix the way it is supposed to, if you get requested visits in a field, set priority for that field to very low, not zero so it can be observed if nothing else can be.
- implement these "standby" methods as time allows
Report from Francisco on filter change limits discussion with camera team and potential impact on proposals (DD in particular
- George - trying to understand if 4 filter changes happen at once, will the heat generation be destructive?
- Impact on deep drilling programs is not horrible but then all allowed filter changes are taken up and Universal maybe heavily impacted.
- May need a better thermal design, will need simulations to tell.: mplement full camera model and see what the impact on (re-optimized) science is.
- what are metrics for srd specs? clearly show that we can or can not meet the srd.
- need to gather these - zejlko and lynne - srd based on Steve's document -
- Cathy to send out Steve's document
- Lynne to send out google doc extension
Refactoring status and developments (if any)
- need to get reviewers feedback on documents
- need to sit with Dave mills to get DDS to work
- vetting telemetry streams
- vetting configs
- Francisco is traveling/moving and sparsely available
- Andy: take one area and do in more detail - API seems good but needs more detail to comment on design - see notes for two areas to work on , e.g. telescope model.
- Michael - detail development as per notes from Sims meeting presentation
- Francisco - has an implementation plan for the filter constraint. It's 2 parts. Telescope model and look ahead. I can describe the needed changes in the telescope model for Michael to consider doing that
Tier 1 runs with v 3_2_1 summary/analysis and discussion of key elements needed for cross simulation comparisons (Kem & all)
- Have been re-run - all runs completed
- first analysis complete; detailed analysis in progress
- main difference is increased downtime - no major issues seen so far
- enigma is now a functioning production machine.
Outstanding MAF issues with current sstarDriver (Cathy).
- new sstarDriver - Cahty/ Kem to check out
- currently use SRD design as benchmark specs, we would like to see how well we do wrt requested visits - Lynne can write a util to get # requested from Config (already in Summary) can be available when running the sstarDriver, and specify requested as benchmark
- consider splitting first loook (srd spec) and engineering (requested) and SRD analysis; obsims wants both. This is a proposed hack for the workshop
Summary of last sprint and new sprint items (preparations for Workshop).
- Sprints will end on Wednesday morning, and beginning on Wednesday afternoon (EOW)
- Reviewers - can we change the JIRA notification that specifies the reviewer; and pair with "in review" and "reviewed".
- how do you find out when you are assigned a reviewer?
- email notification; filter in dashboard; message in Stash;
- Cathy will followup with Frossie for best practices
- Kem - runs with permutations of priority and NES - talk with Lynne and Zeljko to pin down priority
- Lynne working on metric for recovering orbits with visits; rolling cadence will help
- try and down-weight priorities; spend less total fraction of visits - you could have 13%, then 7% for other science.
- ScienceType Tags - need to change in default configs; JIRA needs to hand off the next steps; be better about setting up follow-up JIRA actions
- Add Jiras for workshop presentations etc. from KEm's email
Move to Standup Meetings? next time