[SERVER-79480] Insert measurements into the best-fitting bucket when there are multiple open buckets per metadata Created: 28/Jul/23  Updated: 05/Feb/24  Resolved: 05/Feb/24

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

Type: Task Priority: Major - P3
Reporter: Gregory Wlodarek Assignee: Damian Wasilewicz
Resolution: Won't Do Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Issue Links:
Related
related to SERVER-86222 Complete TODO listed in SERVER-79480 Open
is related to SERVER-81962 Sort unordered bulk inserts by time f... Closed
Assigned Teams:
Storage Execution
Sprint: Execution Team 2023-12-25, Execution Team 2024-01-08, Execution Team 2024-01-22, Execution Team 2024-02-05, Execution Team 2024-02-19
Participants:

 Description   

For a new measurement, the following bucket criteria must be met to consider inserting the measurement into it:

  • The new measurement's metadata matches the bucket metadata.
  • The new measurement's timestamp is the same or later than the last measurement's timestamp in the bucket.

This can return multiple eligible buckets. We pick the bucket where the time difference is the smallest between the new measurement's timestamp and the last measurement's timestamp in each eligible bucket. This approach will minimize the number of open buckets per metadata.



 Comments   
Comment by Gregory Wlodarek [ 05/Feb/24 ]

We're dropping the property that compressed buckets are sorted by time. So we no longer need to support multiple open buckets per metadata. Closing as won't do.

Generated at Thu Feb 08 06:41:05 UTC 2024 using Jira 9.7.1#970001-sha1:2222b88b221c4928ef0de3161136cc90c8356a66.