[SERVER-67141] Use next() instead of seek() in column index stage when possible Created: 08/Jun/22 Updated: 22/Dec/22 Resolved: 22/Dec/22 |
|
| Status: | Closed |
| Project: | Core Server |
| Component/s: | None |
| Affects Version/s: | None |
| Fix Version/s: | None |
| Type: | Task | Priority: | Major - P3 |
| Reporter: | Ian Boros | Assignee: | Dianna Hohensee (Inactive) |
| Resolution: | Won't Do | Votes: | 0 |
| Labels: | pm2646-m5 | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Assigned Teams: |
Query Execution
|
| Sprint: | QE 2022-09-19, QE 2022-10-03 |
| Participants: |
| Description |
|
There are a number of places in the column index stage where we call seek() unconditionally. In the special (and common) case where the cursor is positioned at the RID before the RID we'd seek to, we can call next() instead. Here's one place (for row store fallback). Here's another (used in parent path lookup). |
| Comments |
| Comment by Dianna Hohensee (Inactive) [ 22/Dec/22 ] |
|
Closing as Won't Do. This was filed before column store performance characteristics were better understood. |