Uploaded image for project: 'SAFe Program'
  1. SAFe Program
  2. SP-421

LEAP DD-Calibration Post-CDR Development

Change Owns to Parent OfsSet start and due date...
    XporterXMLWordPrintable

Details

    • Data Processing
    • Hide

      Develop a working prototype, to robustly test our assumptions w.r.t. performance and bottlenecks. 

      A conceptual design of the workflow will be useful to visualise the required inputs (e.g. the required data-products and their order).

      A simple CPU implementation will allow us to explore the workflow hotspots and test variations of the algorithm.

      If the `threading' is as DALiuGE memory drops we can compare with SAGECal under DALiuGE.

      The application will, therefore, have the ability to correct for direction-dependent ionosphere for MWA.

      Show
      Develop a working prototype, to robustly test our assumptions w.r.t. performance and bottlenecks.  A conceptual design of the workflow will be useful to visualise the required inputs (e.g. the required data-products and their order). A simple CPU implementation will allow us to explore the workflow hotspots and test variations of the algorithm. If the `threading' is as DALiuGE memory drops we can compare with SAGECal under DALiuGE. The application will, therefore, have the ability to correct for direction-dependent ionosphere for MWA.
    • Hide

      Outputs:

        * a conceptual design/architecture

        * a CPU example of a parallel LEAP calibration

        * test of performance with this example code, identifying bottlenecks

        * comparison with alternative algorithms such as SAGECal and RTS

      Show
      Outputs:   * a conceptual design/architecture   * a CPU example of a parallel LEAP calibration   * test of performance with this example code, identifying bottlenecks   * comparison with alternative algorithms such as SAGECal and RTS
    • 3
    • 3
    • Team_YANDA
    • Sprint 5
    • Hide

      See comments and associated stories.

      Show
      See comments and associated stories.
    • 9.1
    • Stories Completed, Outcomes Reviewed, Satisfies Acceptance Criteria

    • PI24 - UNCOVERED

    • DP_epic_6 Team_YANDA

    Description

      Low-frequency Excision of the Atmosphere in Parallel LEAP (Rioja, Dodson & Franzen 2018; MN 478, 2337) is a method for evaluating delays for a grid of sources across the sky, from which solutions can be interpolated to arbitrary directions and provided to CSP for beam-forming. It is embarrassingly parallel, and therefore has the potential to run on ionospheric variation timescales, at least for compact configurations.

      In SP-288 we have (I hope) addressed the issues that needed to answered for the CDR, that is the potential of LEAP for real-time beamforming.

      In this feature we will address develop a real-life prototype, and thereby be able to respond to any questions that come back from the CDR.

      The prototyping will be firstly in CPU space, to quickly develop, to debug interactively, and to explore options. The final goal is a GPU version.

      To make this more useful we will work with MWA data. The prototype would then be able to provide LEAP DD-calibration for all MWA data; a pressing current need.

       

       Earmark or a possible stretch/continuation:  a GPU example of a parallel LEAP calibration, with fixes and improvements

      Attachments

        Issue Links

          clones

          Feature - A Feature is a service that fulfills a stakeholder need. Each feature includes a benefit hypothesis and acceptance criteria, and is sized or split as necessary to be delivered by a single Agile Release Train (ART) in a Program Increment (PI). SP-288 Evaluation of LEAP for LOW tied-array beam-forming

          • Not Assigned - Priority is not assigned yet.
          • Done
          relates to

          Feature - A Feature is a service that fulfills a stakeholder need. Each feature includes a benefit hypothesis and acceptance criteria, and is sized or split as necessary to be delivered by a single Agile Release Train (ART) in a Program Increment (PI). SP-437 Assess the performance of MPC consensus approaches

          • Not Assigned - Priority is not assigned yet.
          • Done
          mentioned in

          Page Loading...

          Page Loading...

          Page Loading...

          Structure

            Activity

              People

                r.laing Laing, Robert
                R.Dodson Dodson, Richard
                Votes:
                0 Vote for this issue
                Watchers:
                3 Start watching this issue

                Feature Progress

                  Story Point Burn-up: (100.00%)

                  Feature Estimate: 3.0

                  IssuesStory Points
                  To Do00.0
                  In Progress   00.0
                  Complete823.0
                  Total823.0

                  Dates

                    Created:
                    Updated:
                    Resolved:

                    Structure Helper Panel