lizhimins opened a new issue, #9025: URL: https://github.com/apache/rocketmq/issues/9025
### Is Your Feature Request Related to a Problem? Pop consumption is a new consumption model with the advantages of being lightweight, stateless on the client side, and non-exclusive queueing in load balance. However, the current pop implementation has a large codebase, high disk write traffic when the buffer is closed, and more complex state maintenance when the buffer is enabled. Therefore, we propose a new implementation based on rocksdb storage to elegantly solve the compatibility issues between pop retry topic v1 and v2 formats and the probabilistic retry access problem. Other improvements include full asynchrony operation, optimized cache design, accelerated encoding and decoding, and pop message fairness probleam. Performance testing with a read-write ratio of 1:1 in the same environment compared to the current mainline (Version 5.3.1), the new pop implementation saves 28% of the overall cpu usage of the broker when the buffer is closed, and reduces the overall cpu usage by 4.5% when the Buffer is enabled, with the pop par t showing a reduction of 17%. ### Describe the Solution You'd Like As mentioned above, the proposal aims to eliminate the dependency of pop consumption on scheduled messages, reduce the codebase and implementation complexity, and improve performance to some extent. ### Describe Alternatives You've Considered No ### Additional Context More details: [RIP-59](https://github.com/apache/rocketmq/wiki/%5BRIP%E2%80%9069%5D-Pop-Consumption-Improvement-Based-on-RocksDB) -- 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: commits-unsubscr...@rocketmq.apache.org.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org
