zhijiangW edited a comment on pull request #2: URL: https://github.com/apache/flink-benchmarks/pull/2#issuecomment-678890642
> Do you think this can happen? Sources will produce huge number of records immediately (millions of records per second). Even if checkpoint is triggered just a couple of ms after the source started, the task should be already fully back pressured. If this worries you, maybe there is a way of delaying the first checkpoint? It is also up to how many in-flight buffers we given for network setting. Actually my previous setting of interval 100ms was mainly for delaying the first checkpoint for back pressure, and ignored the delay of intermediate checkpoints. And I also tested to guarantee that there are input & output inflight buffers spilled during checkpointing. I would try out other ways to make sure the back pressure happen. ---------------------------------------------------------------- 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
