Monitor the performance of databases in an instance

This topic describes how to monitor the performance of databases in an instance by using OceanBase Cloud.

Procedure

1. Log in to the OceanBase Cloud console.

2. In the left-side navigation pane, click Instances.

3. On the Instances page, click the name of the target instance to go to the Overview page of the instance.

4. In the left-side navigation pane of the Overview page of the instance, click Metrics.

5. On the Metrics page, the Databases tab automatically appears.

6. By default, the Databases tab shows the monitoring data of all databases in the instance in the last hour.

   • You can select a tenant from the Tenant drop-down list to view its monitoring data. If you select multiple tenants, you can click tenant names in a chart to filter the data to view.
   • You can also specify a time period to filter the data to view.

   The following table describes the performance metrics.

   Monitoring item	Metric	Description
   QPS	sql_all_count	The number of SQL queries processed per second in the database of the tenant.
   Average SQL Statement Processing Time	sql_all_rt	The average time consumed in processing an SQL statement. Unit: ms.
   TPS	transaction_count	The number of transactions requested by the database of the tenant per second. The value is the sum of the following types of requests: INSERT REPLACE UPDATE DELETE
   Transaction response time	transaction_rt	The average time consumed in processing a transaction. Unit: μs.
   Number of Sessions	all_session	The current number of sessions in the database of the tenant.
   Queue time	request_queue_time	The wait time of an SQL query in a queue. Unit: μs.
   Trans Log	trans_commit_log_count	The number of transaction logs committed per second in the database of the tenant.
   Trans Log Size	clog_trans_log_total_size	The total size of transaction logs committed per second in the database of the tenant. Unit: MB.
   IOPS	io_count	The average number of I/O operations per second in the SSStore.
   IORT	io_rt	The average time consumed in each read or write operation in the SSStore. Unit: μs.
   IO Throughput	io_size	The amount of data processed per second in the SSStore. Unit: bytes.
   Transaction Count	transaction_partition_count	The number of transactions committed per second.
   Uptime	uptime	The amount of time when the tenant is in service. Unit: seconds.
   Average time for each transaction log network synchronization	trans_commit_log_sync_rt	The average amount of time consumed on each synchronization of transaction logs over the network. Unit: ms.
   Number of waiting events in a specific state for the database	ob_waiteven_count	The number of wait events in a specific state in the database of the tenant.
   Number of Events in a Specific State	ob_sql_event	The number of execution events in a specific state in the tenant.
   Total number of log disks	ob_tenant_log_disk_total_bytes	The total size of the tenant log disk. Unit: GB.
   Log disk usage	ob_tenant_log_disk_used_bytes	The occupied space of the tenant log disk. Unit: GB.
   Data Usage	ob_tenant_disk_used_percentage	The ratio between the disk space occupied by tenant data and the total disk size. Unit: %.
   Data Volume	ob_tenant_data_size	The volume of tenant data. Unit: GB.
   Binlog Disk Usage	ob_tenant_binlog_disk_used	The size of disk space occupied by binlogs. Unit: GB.