[ https://issues.apache.org/jira/browse/KAFKA-48?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13250982#comment-13250982 ]
Jay Kreps commented on KAFKA-48: -------------------------------- Hey Neha, yes, my hope is to get the patch evaluated as is, and then take another pass at cleaning up the way we handle the satisfaction action as Jun and you requested and try out other approaches to the purgatory data structure asynchronously. That should take these cleanup/polishing items out of the critical path. I like your idea of the dual priority queues, but I need to work through it more to fully understand it. > Implement optional "long poll" support in fetch request > ------------------------------------------------------- > > Key: KAFKA-48 > URL: https://issues.apache.org/jira/browse/KAFKA-48 > Project: Kafka > Issue Type: Bug > Reporter: Jun Rao > Assignee: Jay Kreps > Attachments: KAFKA-48-v2.patch, KAFKA-48-v3.patch, KAFKA-48-v4.patch, > KAFKA-48.patch, kafka-48-v3-to-v4-changes.diff > > > Currently, the fetch request is non-blocking. If there is nothing on the > broker for the consumer to retrieve, the broker simply returns an empty set > to the consumer. This can be inefficient, if you want to ensure low-latency > because you keep polling over and over. We should make a blocking version of > the fetch request so that the fetch request is not returned until the broker > has at least one message for the fetcher or some timeout passes. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira