a2l007 commented on issue #6057: Broker sends sequential requests to the historical for union queries URL: https://github.com/apache/incubator-druid/issues/6057#issuecomment-422912696 @jihoonson Currently the `UnionQueryRunner` executes the query for each datasource sequentially and this is a performance hit for union queries. I'm investigating on how this can be solved. According to my understanding, in the `UnionQueryRunner` all the sequences within the `MergeSequence` would be fetched lazily and so even if I wrap each of these sequences with an AsyncQueryRunner for parallel execution, it wouldn't help because of lazy evaluation. Is my understanding right? Going with my reasoning above, I had suggested the proposal for modifying the `Query` interface for multiple datasources. Please let me know your thoughts.
---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: [email protected] With regards, Apache Git Services --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
