clintropolis commented on a change in pull request #7133: Time Ordering On Scans URL: https://github.com/apache/incubator-druid/pull/7133#discussion_r269838969
########## File path: docs/content/querying/scan-query.md ########## @@ -24,7 +24,16 @@ title: "Scan query" # Scan query -Scan query returns raw Druid rows in streaming mode. +The Scan query returns raw Druid rows in streaming mode. The biggest difference between the Select query and the Scan +query is that the Scan query does not retain all the returned rows in memory before they are returned to the client. +The Select query _will_ retain the rows in memory, causing memory pressure if too many rows are returned. +The Scan query can return all the rows without issuing another pagination query. + +In addition to straightforward usage where a Scan query is issued to the Broker, the Scan query can also be issued Review comment: I think I noticed that the code that hits the historical would maybe explode if not issued a the interval with a `MultipleSpecificSegmentSpec`, but there is an example query immediately after mention that the historicals can be queried that doesn't have this segment spec, and is probably confusing. Querying historicals directly seems like an advanced use case, and should maybe be moved to it's own section near the end of this document to make it clear and not necessarily encouraging it. ---------------------------------------------------------------- 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. 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]
