[SERVER-41244] The WT inMemory storage engine needs an alternative metric to determine cache pressure for the snapshot window size controls Created: 20/May/19  Updated: 06/Jun/19  Resolved: 06/Jun/19

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

Type: Bug Priority: Major - P3
Reporter: Dianna Hohensee (Inactive) Assignee: Dianna Hohensee (Inactive)
Resolution: Won't Fix Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Issue Links:
Backports
Depends
Related
related to SERVER-41560 MongoDB should default the snapshot w... Closed
related to WT-4842 Enhance lookaside score calculation Closed
is related to SERVER-36956 Replace the statistic that dynamicall... Closed
Operating System: ALL
Backport Requested:
v4.2
Sprint: Execution Team 2019-06-03, Execution Team 2019-06-17
Participants:
Linked BF Score: 0

 Description   

The WT inMemory engine does not use a persisted lookaside table for cache overflow, so it needs to look at something other than LAS table writes to determine cache pressure.

SERVER-36956 introduced the problem for the inMemory engine. Perf testing revealed the problem.



 Comments   
Comment by Dianna Hohensee (Inactive) [ 06/Jun/19 ]

I'm closing this as Won't Fix in favor of SERVER-41560. As described in SERVER-41560, we cannot run transactions on an inMemory replica set member in v4.2. Therefore, there is currently no need to maintain snapshot history for the inMemory storage engine. SERVER-41560 will set the window size to 0 for nodes running the inMemory storage engine.

Generated at Thu Feb 08 04:57:12 UTC 2024 using Jira 9.7.1#970001-sha1:2222b88b221c4928ef0de3161136cc90c8356a66.