# Common Monitoring Indicators for TIDB-Server

__*TiDB TiDB-Server Monitoring*__

![left](./student-guide/diagram/MonitoringIndicatorsForTiDBServer_en_001_20220621.png)
-------------------------------------------------------------
Regarding the monitoring method of Prometheus+ Grafana, the monitoring information of TiDB-Server mainly shows information related to performance. In addition, since TiDB connects to components such as TiKV, PD, and client, we can also monitor the connection information. The most commonly used monitoring indicators are:

Statement OPS: The SQL statement executed per second.

Duration: Indicates the average time that the client's SQL statement is sent to the TiDB database and return to the client after the statement is processed. This is a relatively important parameter. In daily operation and maintenance, we first need to monitor this indicator to compare it with the baseline.

Connection Count: Indicates the number of connections to the TiDB database.

Memory Usage: Indicates how much memory is being used, so that we can monitor if an OOM has occurred.

Please note that any monitoring metric must be aware of the timeframe before monitoring, which is the range you are concerned about.