nowinkeyy commented on issue #5184: URL: https://github.com/apache/rocketmq/issues/5184#issuecomment-1279902519
Hi @ShadowySpirits , thank you very much for your reply! > Under normal circumstances, the messages of same queue are ordered in the commitLog. So your improvement is just to execute the dispatcher asynchronously? It seems that the batch write you describe will not work in most cases. The core of this optimization is to process dispatchRequest concurrently, so that different cq can be written at the same time, and even the same cq can be written at limited time (this needs more thought). The proposal has been updated. Now I use batch update commit-position instead of batch write messages. > I don't see anything about recovering from fault in your proposal. eg. Where to start rebuilding cq after broker is down and restart. It is true that I have given less consideration to this aspect, and I prefer to follow the original thinking. **the updated proposal**   -- 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: dev-unsubscr...@rocketmq.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org
