[SERVER-56790] Remove sleeps from the chunk migration recipient catch-up phases Created: 10/May/21  Updated: 06/Dec/22  Resolved: 05/Nov/21

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

Type: Improvement Priority: Major - P3
Reporter: Kaloian Manassiev Assignee: [DO NOT USE] Backlog - Sharding EMEA
Resolution: Won't Do Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Assigned Teams:
Sharding EMEA
Sprint: Sharding EMEA 2021-05-31, Sharding EMEA 2021-06-14, Sharding EMEA 2021-06-28
Participants:

 Description   

The chunk migration recipient side uses a "check opReplicatedEnough" / "sleep" kind of model to wait for replication during the CATCHUP phase. Same thing exists for the STEADY phase, where we check and then sleep.

These should be switched to proper calls to waitForWriteConcern instead.



 Comments   
Comment by Kaloian Manassiev [ 05/Nov/21 ]

From the performance evaluation of migrations of large chunks we have confirmed that the improvements done so far (increasing the batch size and overlapping application of xferMods with fetching them) is sufficient, so we will not be proceeding with this work.

Generated at Thu Feb 08 05:40:12 UTC 2024 using Jira 9.7.1#970001-sha1:2222b88b221c4928ef0de3161136cc90c8356a66.