wang-jiahua commented on PR #10491: URL: https://github.com/apache/rocketmq/pull/10491#issuecomment-4766858269
## Server Benchmark Results (2026-06-22) **Environment**: 8C30G broker, Dragonwell JDK 21, 1KB msg, 128 queues, sync mode, 256 threads, 20s warmup + 30s collect ### Producer-only (steady state, lines 3-5) | Test | TPS | Avg RT | GC | vs Baseline | |------|-----|--------|----|-------------| | Baseline | 217k | 1.18ms | 354 | — | | #10491 | 199k | 1.28ms | 212 | -8.3% TPS (within noise band), -40% GC ✅ | Minor optimizations (TopicMessageType cache, RemotingHelper get+null check, NopStatusListener, metricsValue cache). The -8.3% TPS is at the edge of the noise band (baseline variance is 5-10%). GC reduction (-40%) confirms reduced allocation. These micro-optimizations are not on the Send hot path, so TPS impact is expected to be minimal in production. -- 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]
