yihua opened a new issue, #897: URL: https://github.com/apache/hudi-rs/issues/897
### Feature Description `update_record_index` builds each record-level index shard's log blocks (sorting the shard's keys, encoding its HFile, framing the log block) one shard after another on the task driving the write. That work is CPU-only and independent per shard, but a write that indexes millions of keys runs it on a single core, and it becomes the largest part of a bulk append. Proposed: build the shards' blocks on the blocking pool, a bounded number at a time, and write each finished shard in shard order as it completes. -- 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]
