Uploaded image for project: 'Data Management'
  1. Data Management
  2. DM-19584

Refactor alert distribution system repositories


    • Team:
      Alert Production


      As of DM-17549, there are three repositories related to the LSST alert stream:

      • lsst-dm/sample-avro-alert, which contains:
        • Alert schemas
        • Python library code for working with alerts in Avro format
        • Some trivial examples of alert packets
        • Some miscellaneous Jupyter notebooks, which have been used for benchmarking and creating “postage stamps”
        • A few FITS files for use as postage stamps in sample alerts.
      • lsst-dm/alert_stream, which contains:
        • Python code for sending, filtering and receiving alerts through a Kafka system (both command line tools and supporting libraries)
        • Deployment and configuration scripts for Docker, Docker Swarm, Kubernetes to set up a Kafka system in support of the above.
        • Deployment and configuration scripts to set up a Mirrormaker system.
        • Large volumes of “realistic” simulated alert data (using an old schema)
      • lsst-dm/filtering-blueprints, which contains “Place to keep notes/snapshots relevant to alert stream processing (filtering) framework prototyping.”.

      This is confusing: please rationalise it! In particular, it should be easy for end users to access alert schemas without downloading irrelevant data or code, and to access libraries for working with alerts without being deluged by Docker/Swarm/Kafka/etc configuration.



          There are no comments yet on this issue.


            • Assignee:
              swinbank John Swinbank
              swinbank John Swinbank
              John Swinbank
            • Votes:
              0 Vote for this issue
              1 Start watching this issue


              • Created:

                Summary Panel