himanshug commented on issue #6088: Scan query: time-ordering URL: https://github.com/apache/incubator-druid/issues/6088#issuecomment-411139043 Implementing (3) makes sense to me as well with one additional change that if limit was L then it would work like below. try finding latest L (or less) in first "hour", say some x < L found (write these x rows to response out) then try finding latest (L-x) in second "hour"... ... keep going till you either find L rows or have been through all segments. the "hour" could be "minute", "day" etc ... it is either provided by the user in the query or automatically determined (e.g. use segment granularity). By breaking it down, I'm basically hoping there wouldn't 100s of segments in one "hour" window leading to the problem mentioned in (4).
---------------------------------------------------------------- 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: us...@infra.apache.org With regards, Apache Git Services --------------------------------------------------------------------- To unsubscribe, e-mail: commits-unsubscr...@druid.apache.org For additional commands, e-mail: commits-h...@druid.apache.org
