vagetablechicken commented on issue #2780: OlapTableSink::send is low efficient? URL: https://github.com/apache/incubator-doris/issues/2780#issuecomment-600976814 > The new non-blocking approach looks good. It should improve the overall load performance particularly in case where you have some slower receiver nodes. > > BTW you should cap the total number of pending batches in order to control the memory usage. There's no need to limit mem in sink node. When we create a new RowBatch in NodeChannel, we use sink node's mem_tracker. As we know, the sink node & the scan node has the same ancestor mem_tracker(query mem tracker， default 2GB), so the mem limit is a matter for scan node.
---------------------------------------------------------------- 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: us...@infra.apache.org With regards, Apache Git Services --------------------------------------------------------------------- To unsubscribe, e-mail: commits-unsubscr...@doris.apache.org For additional commands, e-mail: commits-h...@doris.apache.org
