Multi-dimensional cluster metric details

View multi-dimensional cluster monitoring data

1. Log on to the OceanBase Management Console .

2. In the navigation pane on the left, click Instance List.

3. In the instance list, find the target instance and click the instance name to open the Instance Workbench.

4. In the navigation pane on the left of the workbench page, click Performance Monitoring.

5. Click the Multi-dimensional Cluster Metrics tab.

6. By default, the multi-dimensional cluster metrics monitoring page displays the monitoring data for all tenants for the past hour.

   • From the Metric Dimension drop-down list, select a dimension to view its monitoring data. The available dimensions are cluster, tenant, replica, and node.

   • You can filter by a custom time range to view data for that period.

   

   The following monitoring metrics are available:

   Metric Item	Metric Name	Description
   QPS	sql_all_count	The number of SQL statements processed per second by the tenant database. Unit: times/s.
   Average SQL statement processing time	sql_all_rt	The average processing time of an SQL statement. Unit: ms.
   TPS	transaction_count	The number of transactions requested per second from the tenant database. Unit: times/s. This metric is the sum of the following request types: Insert Replace Update Delete
   Average server-side transaction processing time	transaction_rt	The average processing time of a transaction. Unit: μs.
   Number of sessions	all_session	The current number of sessions in the tenant database.
   Request queue wait time	request_queue_time	The time an SQL statement waits in the queue. Unit: μs.
   Number of transaction logs committed per second	trans_commit_log_count	The number of transaction logs committed per second by the tenant database. Unit: times/s.
   Size of transaction logs committed per second	clog_trans_log_total_size	The size of transaction logs committed per second by the tenant database. Unit: MB.
   Average I/O operations per second	io_count	The number of I/O operations per second for SSStore. Unit: times/s.
   I/O time	io_rt	The average time for each SSStore read/write operation. Unit: μs.
   I/O throughput	io_size	The data volume processed per second by SSStore. Unit: byte.
   Number of transactions	transaction_partition_count	The number of transactions per second. Unit: times/s.
   Available time	uptime(s):	The available time. Unit: s.
   Average network synchronization time per transaction log	trans_commit_log_sync_rt	The average network synchronization time for each transaction log. Unit: ms.
   Number of wait events for the database in a specific state	ob_waiteven_count	The number of wait events for the database in a specific state.
   Number of execution events for the database in a specific state	ob_sql_event	The number of execution events for the database in a specific state.
   Total log disk size	ob_tenant_log_disk_total_bytes	The total size of the log disk. Unit: GB.
   Used log disk size	ob_tenant_log_disk_used_bytes	The used size of the log disk. Unit: GB.
   Data usage	ob_tenant_server_required_size	The data usage. Unit: GB.
   Data volume	ob_tenant_server_data_size	The data volume. Unit: GB.
   Binlog disk usage	ob_tenant_binlog_disk_used	The disk usage of binlogs. Unit: GB.