List view

• Windowing v1

  The goal is to make SQLFlow a viable alternative to Flink (or comparable stateful solution). This means SQLFlow needs robust windowing support. Support should include persistent windows, as well as a variety of common window types, including tumbling, and hopping. Windowing needs to be well documented and operationally friendly. This milestone builds out a cohesive initial windowing capability.

  No due date

  •0/4 issues closed
  0% complete4 open0 closed

• Connectivity Ecosystem v1

  This milestone focuses on building out SQLFlow connectivity allowing it to ingest data from a variety of sources and sink data to a variety of useful places. SQLFlow should add support for generic support of any duckdb sink, to minimize the custom code necessary to add new data sinks.

  No due date

  •3/5 issues closed
  60% complete2 open3 closed

• Operational Readiness v1

  Operational Readiness Focuses on running SQFLow at scale for long periods of time. This includes things like: - Monitoring - Alerting Recommendations - Debugging / Log - Ensuring stack traces are dumped on error - Feedback on invalid configurations When this is complete SQLFlow should be observable and ready to run in a production-like environment powering real workloads.

  No due date

  •7/11 issues closed
  63% complete4 open7 closed

• v1.0.0

  Stable release, SQLFlow has enough features to confidently run in production on production workloads.

  No due date

  •8/9 issues closed
  88% complete1 open8 closed