danny0405 commented on PR #5629: URL: https://github.com/apache/hudi/pull/5629#issuecomment-1202059975
> I don't think we are aligned on this one: Hudi is and will be staying engine-neutral project. However for the top-of-the-line performance on _any_ engine (to stay competitive with other formats) we _have to_ use engine-specific representations (think `Row`, `RowData`, `ArrayWritable`, Arrow, etc). There's just no other way I do have some different thoughts, performance is on first priority if it is critical, say Hudi performs bad on some benchmark, but for the long run, i do think as a storage we should have our own data structures/reader writers/schema like every storage engine do, looks like how easy it is to a new engine to integrate with Iceberg and how hard it is for Hudi, the ease to integrate is important for the ecosystem especially as a `format`. -- 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]
