Nieal-Yang commented on pull request #2375: URL: https://github.com/apache/hudi/pull/2375#issuecomment-754359543
> > Hi @garyli1019. Maybe I think the current implementation is OK. Beacause even in streaming job, we need to accumulate batch records in memory during the check-point cycle and upsert data into hudi-table when check-point triggers. WDYT? > > @Nieal-Yang sorry about the late reply. IMO The batch mode will work but not fully taking the advantage of flink. But we can optimize this step by step. To ensure this PR will work, would you add some unit tests to this PR? okay. This index has been used in production in my company. We will optimize it gradually ---------------------------------------------------------------- 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]
