regadas commented on PR #33648: URL: https://github.com/apache/beam/pull/33648#issuecomment-2605923347
Hello @chamikaramj Thanks glad to help 👍 >To clarify, do the performance gains here come from eliminating per-element "TableIdentifier.parse(element.getKey())" calls or something else ? yeah ... but with the proposal on https://github.com/apache/beam/pull/33293 things got a bit worse since parsing Json per element becomes very expensive; > If pref gains are similar, I think it's preferable to not break update compatibility :) agree, tbh haven't done benchmark yet, let me get back to you on this > Also, as a side note, we heavily discourage using Java serialization for coders. for good reasons 😄 , need to double check where TableIdentifierCoder is not used -- 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]
