This message was deleted.

Yes, Pulsar exposes the metrics via a Prometheus endpoint. You can scrape those metrics and feed them into Grafana

@David K thank you so much for your great response, the two tools are already running in the k8 cluster just trying to determinng the accurate metric that plots this type of result

2023-11-30T21:37:47,071+0000 [main] INFO  org.apache.pulsar.testclient.PerformanceProducer - Throughput produced:    9808 msg ---    499.9 msg/s ---      3.9 Mbit/s  --- failure      0.0 msg/s --- Latency: mean:  16.932 ms - med:  16.192 - 95pct:  24.683 - 99pct:  35.888 - 99.9pct:  65.253 - 99.99pct:  71.568 - Max:  73.668

2023-11-30T21:37:57,101+0000 [Thread-0] INFO  org.apache.pulsar.testclient.PerformanceProducer - Aggregated throughput stats --- 10000 records sent --- 331.788 msg/s --- 2.592 Mbit/s 
2023-11-30T21:37:57,117+0000 [Thread-0] INFO  org.apache.pulsar.testclient.PerformanceProducer - Aggregated latency stats --- Latency: mean:  19.296 ms - med:  16.274 - 95pct:  27.028 - 99pct: 126.585 - 99.9pct: 294.021 - 99.99pct: 304.353 - 99.999pct: 306.429 - Max: 306.429

Most of those would fall under “topic metrics”.

thank you @David K you are the best