danny0405 commented on PR #13265: URL: https://github.com/apache/hudi/pull/13265#issuecomment-2864847033
> The algorithm’s reliance on multiplicative offsets (partitionIndex = hash % parallelism * bucketNum) and modulo arithmetic amplifies skew when bucketNum and parallelism share common factors. This design flaw inherently limits the effective parallelism So there is no way to improve the local algorithm anyway? -- 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]
