Uploaded image for project: 'Core Server'
  1. Core Server
  2. SERVER-22460

Logic for applying distinct scan and count scan optimizations should be moved to data access planner

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

      getExecutorDistinct() and getExecutorCount() have special performance-related transformations that they attempt to apply to each query solution tree returned from the query planner (both attempt to rewrite the query solution tree to contain DistinctNode or CountScan nodes instead of IndexScanNodes). These optimizations should instead be moved to the access planner; doing so will allow for unit testing of this logic.

            Assignee:
            backlog-query-optimization [DO NOT USE] Backlog - Query Optimization
            Reporter:
            rassi J Rassi
            Votes:
            0 Vote for this issue
            Watchers:
            5 Start watching this issue

              Created:
              Updated: