[SERVER-66262] Consider refactoring how we dedupe index bounds in SBE INLJ Created: 05/May/22 Updated: 16/Mar/23 Resolved: 16/Mar/23 |
|
| Status: | Closed |
| Project: | Core Server |
| Component/s: | None |
| Affects Version/s: | None |
| Fix Version/s: | None |
| Type: | Task | Priority: | Minor - P4 |
| Reporter: | Irina Yatsenko (Inactive) | Assignee: | Backlog - Query Execution |
| Resolution: | Won't Do | Votes: | 0 |
| Labels: | None | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Assigned Teams: |
Query Execution
|
| Participants: | |
| Story Points: | 5 |
| Description |
|
Currently, the plan produces an ArraySet with local key values, including arrays, unwinds it and then creates an index bound for each unique value and possibly for the first element of each array. The latter might create a duplicated index seek and we avoid it by filtering this branch out in case of key value duplication. This is brittle and probably not the most efficient way. We should consider refactoring the plan to have more robust deduplication of index bounds. |
| Comments |
| Comment by Mihai Andrei [ 16/Mar/23 ] |
|
Going to kick this out of the project and close as won't do; feel free to reopen/flag for triage if you think this work is worth doing! |