[
https://issues.apache.org/jira/browse/FLINK-36909?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
xiaochen.zhou updated FLINK-36909:
----------------------------------
Description: In StreamPatternSubscriber mode, our current approach is to
first fetch metadata for all streams, then create all streams, and subsequently
apply filters. We can optimize this process by initially filtering stream
identifiers, then fetching metadata and creating streams only for the filtered
ones, thereby enhancing performance and reducing resource consumption.
> Reduce Stream Creation in StreamPatternSubscriber
> -------------------------------------------------
>
> Key: FLINK-36909
> URL: https://issues.apache.org/jira/browse/FLINK-36909
> Project: Flink
> Issue Type: Improvement
> Components: Connectors / Kafka
> Reporter: xiaochen.zhou
> Priority: Minor
> Labels: pull-request-available
>
> In StreamPatternSubscriber mode, our current approach is to first fetch
> metadata for all streams, then create all streams, and subsequently apply
> filters. We can optimize this process by initially filtering stream
> identifiers, then fetching metadata and creating streams only for the
> filtered ones, thereby enhancing performance and reducing resource
> consumption.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
