[SERVER-26778] Improve the speed of incremental chunk metadata refresh Created: 25/Oct/16  Updated: 08/Feb/18  Resolved: 08/Feb/18

Status: Closed
Project: Core Server
Component/s: Sharding
Affects Version/s: 3.2.10, 3.4.0-rc1
Fix Version/s: None

Type: Improvement Priority: Major - P3
Reporter: Daniel Pasette (Inactive) Assignee: Kaloian Manassiev
Resolution: Duplicate Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Issue Links:
Duplicate
duplicates SERVER-32526 Use KeyString for the ChunkManager's ... Closed
Related
related to SERVER-25652 Slow chunk migrations when there are ... Closed
Participants:
Case:

 Description   

Initial chunk load, which requires a table scan, is 1.1 seconds:

2016-10-24T16:18:01.297+0000 I SHARDING [Balancer] ChunkManager: time to load chunks for buildlogs.logs: 1133ms sequenceNumber: 2 version: 157970|1||576339e709168e06c8eb8097 base

Subsequent refreshes, which require a very fast query (0 ms) still take over 250ms:

2016-10-24T16:33:41.683+0000 I SHARDING [Balancer] ChunkManager: time to load chunks for buildlogs.logs: 264ms sequenceNumber: 3 version: 157970|4||576339e709168e06c8eb8097 based

This issue is to investigate the issue.


Generated at Thu Feb 08 04:13:11 UTC 2024 using Jira 9.7.1#970001-sha1:2222b88b221c4928ef0de3161136cc90c8356a66.