Instance monitoring

Updated at:
Copy as MD

PolarDB-X 1.0 provides instance monitoring so you can track CPU, memory, network, and query performance in real time. This topic describes how to view monitoring data on the Instance Monitoring page in the PolarDB for Xscale console.

Monitoring metrics

PolarDB-X 1.0 groups metrics into two categories:

  • Resource: Infrastructure-level metrics (CPU, memory, network). Collected every 1 minute.

  • Engine: Database engine metrics (query throughput, response time, connections, threads). Collected every 5 seconds.

Monitoring data is retained for seven days.

Resource metrics

Metric Parameter Unit Description
CPU cpu % Average CPU utilization of the compute node.
Memory old % Memory usage of the compute node. Fluctuations are normal.
Network (inbound) netin Kbit/s Total inbound traffic of the compute node. Inbound traffic is generated when the storage node returns data to the compute node.
Network (outbound) netout Kbit/s Total outbound traffic of the compute node. Outbound traffic is generated when the compute node sends physical SQL statements to the storage node, or when the compute node returns data to applications.

Engine metrics

Metric Parameter Description
Logical QPS lqps Total number of SQL statements processed by the instance per second.
Physical QPS pqps Total number of SQL operations sent from the compute node to the storage node per second.
Logical RT lrt Average response time (RT) for each SQL statement processed by the instance.
Physical RT prt Average RT for SQL statements sent from the compute node to the storage node.
Active connections con Total number of connections between an application and the instance.
Active threads thread Number of threads executing SQL statements in the instance.

View instance monitoring data

  1. Log on to the PolarDB for Xscale console.

  2. In the top navigation bar, select the region where your instance is located.

  3. On the Instances page, find the instance and click its instance ID.

  4. In the left-side navigation pane, choose Monitoring and Alerting > Instance Monitoring.

  5. On the Instance Monitoring page, set the following parameters.

    Parameter Description
    Monitoring Index The metric category. Select Resource or Engine.
    Monitoring Metric The monitoring metrics of the instance. Select the specific metrics to display.
    Query Time The time range for monitoring data. Options: 1 hour, 6 hours, 12 hours, one day, or one week. To set a custom range, specify a start time and end time. The minimum range is 1 minute and the maximum range is one week.