yihua opened a new issue, #900: URL: https://github.com/apache/hudi-rs/issues/900
### Feature Description Before the record-level index shards are built, `update_record_index` walks every entry on the writing task to route it to its shard and encode its Avro payload, so a write that indexes millions of keys spends a serial pass encoding them before the parallel shard build starts. Only the routing has to happen up front. Proposed: route entries to shards by index on the writing task, and let each shard's task encode its own entries. -- 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]
