austince commented on pull request #15497: URL: https://github.com/apache/flink/pull/15497#issuecomment-815241115
So @zentol, I think we have three options: 1. do not support this case in the adaptive scheduler (882e7fd + b412a4b) 2. support this on a "best-effort" basis, applying the same rules as Reactive mode that max parallelism cannot be scaled down (eaf1f34 + f39c534) 3. support this use case fully by changing max parallelism after the fact (eaf1f34 (?) + ...) Maybe there are options that I'm missing. But from this list, I'd choose 1. What do you think? -- 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]
