[SERVER-70241] [CQF] Implement "auto mode" for cardinality estimation Created: 05/Oct/22  Updated: 08/Feb/23

Status: Backlog
Project: Core Server
Component/s: None
Affects Version/s: None
Fix Version/s: None

Type: Improvement Priority: Major - P3
Reporter: Svilen Mihaylov (Inactive) Assignee: Backlog - Query Optimization
Resolution: Unresolved Votes: 0
Labels: bonsai-ce
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Issue Links:
Depends
depends on SERVER-68490 Integrate histogram estimation with t... Closed
Assigned Teams:
Query Optimization
Participants:

 Description   

This is regarding the flag "kInternalCascadesOptimizerDisableScanDefault".

 

Currently we have 3 options: heuristic, sampling, and histograms. The sampling is currently ignored for small sizes while the histogram is not. The reasoning behind is that it may be too slow given the relative collection size.

 

The goal of this ticket is to:

  1. Add "auto" CE estimator mode which will select the best estimator (contingent on current efforts to implement estimation based on histograms)
  2. Allow forcing of estimation strategies (independent on the metadata).

Generated at Thu Feb 08 06:15:39 UTC 2024 using Jira 9.7.1#970001-sha1:2222b88b221c4928ef0de3161136cc90c8356a66.