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#

Agenda & Notes

The primary topic will be a discussion of Francisco’s refactoring document and how to move forward (see his email of 1/16/2015).

  1. Identify the components and what their roles are (e.g. telemetry, control, weather, history of visits etc). Francisco's document does most of this
  2. Identify the API for each of these components (ie what is passed into and out of the scheduler and each component that generates telemetry data). I think this is the next step we need to complete as that will help define the class structure and the timescales on which we update the system
  3. Define what time keeping we need and how we will pass time into the scheduler at different rates (ie operations and simulations)
  4. Define what sort of data structures we will need to work with (e.g. healpix maps) within the scheduler

Refactoring plan

  • talking about freezing the code and work on a re-write.
  • refactoring takes prioritiy over pex-config
  • use v3.2.1 to produce runs for workshop and community but future efforts will be towards redesign and rewrite to interface with DDS and make it easier to add features
  • swap in scheduling algorithms
  • 3 or more visits in the universal cadence - franscico do do in a day or so??  or just not do

Review documents from Francisco

  • control system controls system
  • scheduler picks next field
  • simulator simulates the conditions and visits and reports to the scheduler.
  • review of block diagrams for scheduler and simulator

Identify flow of information

  • which part has to generate which information (data map, single number, etc)
  • what will change on what timescales
  • how do we record this?  work with Michael work on Feb 2

Tasks for Feb 2

  • itemize what information is passed where and how is it specified (understand how to do dev work)
  • DDS as a tool?  suitability
  • Simulator re-write?
  • any changes to v3.2.1 - freeze?  multi visits? other?
  • Michael and Francisco- review the life cycle of a visit - map the transfer of information into the diagrams.
  • SchedulerOpSimInterfaces-1.1.pdf

 

If Andy is available, we will continue arguing about pexconfig implementation, parameter tracking and docker implementation.

 

Kem and Cathy will give an update on opsim v3.2.1 and maf on production machines.

 

Update on the new (actually current) workflow implementation.

 

Alignment of sprints across Simulations and creation of single JIRA project for Simulations discussion.

  • No labels