csviri commented on PR #281: URL: https://github.com/apache/spark-kubernetes-operator/pull/281#issuecomment-3070122128
> Thank you for the details. Can we have some numbers for this claim, @csviri ? > > > Having additional informer might effect performance, memory consumption, also maintains an additiona websocket connection to Kubernetes API. Unfortunately, I don't have any at hand now. Just to shed a light on this: - performance wise the reconciler will start reconiling the first resource when all the informers are synced (this includes the informer of the primary, thus listed all the resources of target resources into the cache) - additional infomer means it caches the target resources, in this case basically the informer was redundant, so had the resources twice in memory - additional informer means an additional websocker connection maintained continuously to K8S API server. Will work on the number within JOSDK in close future. cc @shawkins do you happen to have some numbers / metrics regarding informers? thank you -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
