slinkydeveloper commented on pull request #17520: URL: https://github.com/apache/flink/pull/17520#issuecomment-947418884
> The biggest concern is that StreamFormatAdapter.Reader#readBatch stores all results in a batch in heap memory. This is bad because avro is a format which supports compression. You'll never know how much data will be stuffed into heap memory after inflation. > But Avro has own batch and compression logical. This conflict does lead to poor work in some places. Can you please elaborate? I don't understand where in this code changes we exploit the avro batch and compression logic. From the code in this PR, I still see we invoke the data file iterator record per record https://github.com/apache/flink/pull/17520/files#diff-07c21eca7aca500ba4675ecd3ace539cf31d69797fd254b7b914d22198a789baR157. -- 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]
