TiDB Monitoring Metrics Explanation Issue - Can you explain the terms for the 80th, 95th, 99th, and 99.9th percentiles? It seems like I understand but also don't quite understand

Note:
This topic has been translated from a Chinese forum by GPT and might contain errors.

Original topic: tidb-监控指标解释问题-能否解释一下80,95,99,999四张百分比的名词,好像理解又不太理解

| username: yulei7633

The image you provided is not accessible. Please provide the text content you need translated.

| username: yulei7633 | Original post link

Looking at the above image, the 80th percentile duration is very small, but the 999th percentile is high. Can it be understood that if there are 100 SQL statements, 80 of them have low duration, while the remaining 20 have relatively high duration?

| username: zhanggame1 | Original post link

How is P99 calculated in Prometheus? - Zhihu (zhihu.com)

Refer to this.

| username: DBRE | Original post link

The figure represents the execution time of 80%/95%/99%/99.9% of SQL requests within X ms.

| username: 小龙虾爱大龙虾 | Original post link

The expert above is correct. The chart represents the execution time of 80%/95%/99%/99.9% of SQL requests within X ms. In Prometheus, this is recorded using a histogram type. You can check the data collected by the metrics of the TiDB component by using the command: curl http://{tidb-server-ip}:10080/metrics.

| username: yulei7633 | Original post link

Many thanks

| username: yulei7633 | Original post link

Thank you very much.

| username: dba远航 | Original post link

#80, 95, 99, and 999 correspond to the respective performance baselines.

| username: yulei7633 | Original post link

Thank you very much!

| username: system | Original post link

This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.