-
Type: Bug
-
Resolution: Works as Designed
-
Priority: Major - P3
-
None
-
Affects Version/s: None
-
Component/s: None
How frequently does the bug occur?
Sometimes
Description
Hi Realm team,
I am filing this ticket as requested by @ianpward over email. This is an inquiry about realm performance around Mongo queries, and is a reference for your investigation. I do not believe this issue is specific to realm-js.
App Services https://realm.mongodb.com/groups/632e05374a332334728f1257/apps/638f4bbacfca527011d14f65/dashboard
App ID: stashpad-bbbqw
Our Atlas cluster frequently scales up and down due to high CPU usage. There appears to be a high number of disk scans as a part of realm syncing between devices and the app services data.
In general, the scaling happens more frequently than we'd expect for having only a couple of hundred concurrent users at the most.
As a specific example, here is a spike that caused us to scale from M20 -> M30.
Though there was a relatively small amount of data transferred according to the app services app.
We were not initially concerned by the amount of scaling but were contacted by a MongoDB customer success rep who suggested that scaling this frequently is bad, and that our cluster was having a high number of disk scans and in memory sorts.
Please let me know if I can provide any more information to help further this investigation.
Stacktrace & log output
No response
Can you reproduce the bug?
Sometimes
Reproduction Steps
Normal use of the Realm JS SDK in both a react native app, and an electron app.
Version
11.10.1
What services are you using?
Atlas Device Sync
Are you using encryption?
No
Platform OS and version(s)
N/A
Build environment
No response
Cocoapods version
No response