GGraziadei commented on PR #8593: URL: https://github.com/apache/storm/pull/8593#issuecomment-4827545023
Thank you both for the review and the insightful discussion! First of all, my apologies for not replying explicitly earlier, I have already merged the main branch and resolved the open conflicts. Tomorrow, I will update the documentation to improve the presentation of the benchmark results, framing them in a more neutral way as suggested (highlighting the directional latency improvement under moderate load and clarifying that tuple ordering is not guaranteed). Regarding the mixing of the control plane and data plane on the same queues: I completely agree. Under heavy load, this buffering delay will inevitably make the jitter/feedback mechanism less responsive. Tomorrow, I will open a dedicated follow-up issue to address this. My initial thought is that we could resolve this by decoupling the queues and separating anchored/unanchored traffic, but it will definitely be useful to discuss with and evaluate all possible options within that ad-hoc issue. Thanks again for catching this! -- 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]
