rangareddy commented on issue #17122: URL: https://github.com/apache/hudi/issues/17122#issuecomment-5366098842
This issue was reviewed as part of the JIRA-migrated backlog triage (HUDI-9662). **Findings: still open.** The need is concrete: bloom-filter behaviour currently cannot be varied per metadata partition, so FILES, COLUMN_STATS and RECORD_INDEX are forced to share settings despite very different key cardinalities and lookup patterns. Worth noting the index-version machinery added since (`HoodieIndexVersion.getCurrentVersion` returning V1/V2 per `MetadataPartitionType`) is a natural place for per-partition behaviour to hang off, so this may be cheaper now than when filed. **One thing worth flagging to avoid a mis-triage:** the PR linked in the description, #13563 (`[HUDI-9628] Add bloom filtering to prune keys prior to lookup in MDT`), **is merged** (2025-07-30). But that is the change this ticket was filed as a follow-up to - not a fix for it. Nothing indicates the follow-up work itself has landed. Keeping this open. -- 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]
