[SERVER-84944] Last point TS Opt M3: TSBS insert performance experiment with doubled indexes Created: 04/Mar/22  Updated: 12/Jan/24  Resolved: 07/Mar/22

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

Type: Task Priority: Major - P3
Reporter: Steve Tarzia Assignee: Alya Berciu
Resolution: Done Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Attachments: PNG File tsbs_load_fast.png     PNG File tsbs_load_fast_3node.png     PNG File tsbs_load_manual.png     PNG File tsbs_load_manual_3node.png     PNG File tsbs_load_slow.png     PNG File tsbs_load_slow_3node.png    
Sprint: QO 2022-03-21
Participants:

 Description   

We need to measure the performance impact of our proposed solution to the reverse DISTINCT_SCAN for $sort+$group+$last.

There's no need to merge anything, but we should try updating the TSBS insert tests to manually include indexes on the missing {meta, min, max} or {meta, max, min} bucket index for each defined {meta, time} index.  After doubling the number of indexes, how is performance affected?  If there's not much change, we can proceed confidently with the plan to double these indexes.


Generated at Thu Feb 08 06:56:25 UTC 2024 using Jira 9.7.1#970001-sha1:2222b88b221c4928ef0de3161136cc90c8356a66.