Integrate join optimization into pipeline optimization

XMLWordPrintableJSON

    • Type: Task
    • Resolution: Unresolved
    • Priority: Major - P3
    • None
    • Affects Version/s: None
    • Component/s: None
    • Query Optimization
    • None
    • None
    • None
    • None
    • None
    • None
    • None

      This ticket represents the culmination of the first milestone. This ticket will integrate the pieces that we're previously implemented together.

      • Consult feature flag in run_aggregate and branch into the new join reordering module
      • Ensure we are in situation where join reordering applies (no sharded collections, all collection co-located, etc.)
      • Attempt to extract join graph. Fallback to existing code upon failure.
      • Invoke single table access path selection for each referenced collection
      • Construct single physical plan (QuerySolution) with a pseudo random traversal of the join graph.
      • Invoke stage builders to construct SBE executable plan
      • Attach suffix DocumentSources that were not pushed down

            Assignee:
            Alya Berciu
            Reporter:
            Ben Shteinfeld
            Votes:
            0 Vote for this issue
            Watchers:
            5 Start watching this issue

              Created:
              Updated: