You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Even after #1046 was fixed, opening a synchronized Realm in Realm Kotlin is still slower than on Realm Java and we need to investigate why.
Right now my guess is that either our schema-generation code or spinning up background notifiers is slowing things down, but this will need to be investigated. It might be investigated to #908
The project that shows this issue has a difference of 3.6 seconds, 4.4 seconds on Java, and 8 seconds on Kotlin. This includes downloading the Realm data from the server. See https://jira.mongodb.org/browse/HELP-37904 for more information.
4.4 seconds is a lot of time though, I would be surprised if it took that long to spin up a few thread pools and reading the schema. So maybe there is something else we are missing.
This ticket just covers discovering what the root cause of the problem is.
The text was updated successfully, but these errors were encountered:
I believe @clementetb was looking into this at some point and we measured this at some point and speculated if it would make sense to find a way to bulk-send all schema information through JNI rather than now where we go through the JNI layer for each property.
We also need to look into Realm Java and see if adopting the same pattern in Kotlin would make sense.
Even after #1046 was fixed, opening a synchronized Realm in Realm Kotlin is still slower than on Realm Java and we need to investigate why.
Right now my guess is that either our schema-generation code or spinning up background notifiers is slowing things down, but this will need to be investigated. It might be investigated to #908
The project that shows this issue has a difference of 3.6 seconds, 4.4 seconds on Java, and 8 seconds on Kotlin. This includes downloading the Realm data from the server. See https://jira.mongodb.org/browse/HELP-37904 for more information.
4.4 seconds is a lot of time though, I would be surprised if it took that long to spin up a few thread pools and reading the schema. So maybe there is something else we are missing.
This ticket just covers discovering what the root cause of the problem is.
The text was updated successfully, but these errors were encountered: