echooymxq opened a new issue, #10432: URL: https://github.com/apache/rocketmq/issues/10432
### Before Creating the Enhancement Request - [x] I have confirmed that this should be classified as an enhancement rather than a bug/feature. ### Summary Currently, when RocketMQ uses RocksDB-based ConsumeQueue, the default read path is controlled by `iteratorWhenUseRocksdbConsumeQueue`, which is enabled by default. However, this iterator is a RocketMQ-level reusable iterator. Internally, it still calls `RocksDBConsumeQueueTable#rangeQuery`, which builds N RocksDB keys for consecutive ConsumeQueue offsets and then calls `multiGet`. This enhancement proposes optimizing the RocksDB ConsumeQueue sequential read path by using a bounded `RocksIterator` scan for consecutive offsets, while preserving the existing external ConsumeQueue iterator semantics. ### Motivation Using RocksDB's native iterator scan for continuous offsets to improve performance. ### Describe the Solution You'd Like Use a bounded RocksIterator scan in RocksDBConsumeQueueTable#rangeQuery to read continuous offsets while keeping existing ConsumeQueue iterator semantics unchanged. ### Describe Alternatives You've Considered No ### Additional Context _No response_ -- 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]
