beliefer commented on PR #36405: URL: https://github.com/apache/spark/pull/36405#issuecomment-1121782479
> Please elaborate on the idea more. I don't get how to avoid OOM by not pushing down the limit. The tasks on executors will pull all the result set from data source anyway, isn't it? Then tasks will send the local limit result to the driver and do a final limit. I reconsidered the idea. As you said, it seems we can not avoid OOM if the `LIMIT` value if too large. Maybe, we could add document as @huaxingao said. -- 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. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
