All Products
Search
Document Center

PolarDB:Performance monitoring

Last Updated:May 13, 2025

PolarDB console provides various performance metrics and monitoring at intervals of several seconds, enabling you to check your cluster status and quickly identify operational issues using detailed monitoring data.

Performance monitoring

  1. Log on to the PolarDB console, select the region where your cluster is deployed, and click the cluster ID in the Clusters list to go to the details page.

  2. In the left-side navigation pane, click Performance Monitoring.

  3. You can view monitoring information on the Cluster, Compute Node, Advanced Monitoring, or Performance Events tab based on your business needs.

    • Cluster performance monitoring: Click the Cluster tab, set the time period on the right, and click OK.

    • Compute node performance monitoring: Click the Compute Node tab, select a node and set the time period on the right, and click OK.

    • Advanced monitoring: Click the Advanced Monitoring tab, select a node and set the time period below, and click OK.

    • Performance events: Click the Performance Events tab, select a node and set the time period below, and click OK.

Metrics

Category

Metric

Description

Cluster

CPU utilization

Displays the CPU utilization of each node.

Memory usage

Displays the memory usage of each node.

Storage capacity

Displays the storage occupied by data files, logs, and write-ahead logs (WALs), and the temporary storage.

Compute node

CPU utilization

Displays the CPU utilization of a selected node.

PCU CPU utilization

Displays the CPU utilization of a selected node in PCU units.

Note
  • You can use this metric to view specific CPU utilization only for serverless clusters.

  • PCU stands for PolarDB Capacity Unit. 1 PCU is approximately equal to the capability of 1 core and 2 GB of memory. PCU is the management unit for resource scalability in PolarDB serverless clusters, with a minimum step of 0.5 PCUs.

Memory usage

Displays the memory usage of the selected node.

Connections

Displays the total number of current connections, the number of active connections, and the number of idle connections on the selected node.

TPS

Displays information about the transactions that are performed per second on the selected node. The information includes the number of committed, locked, and rolled back transactions per second.

Scanned rows

Displays the number of rows that are inserted, read, updated, deleted, and returned per second on the selected node.

Maximum database age

Displays the difference between the ID of the first transaction and the ID of the most recent transaction in the database.

I/O throughput

Displays the total I/O throughput, read I/O throughput, and write I/O throughput of the selected node.

IOPS

Displays the total IOPS, read IOPS, and write IOPS of the selected node.

Cache

Displays the number of cache reads per second and the number of disk reads per second on the selected node.

Cache hit ratio

Displays the cache hit ratio of the selected node.

Temporary files

Displays the number and the total size of temporary files on the selected node.

Advanced monitoring

PolarDB CPU utilization/memory usage

Displays the CPU usage and memory usage of the selected node.

TPS/QPS

Displays the TPS and QPS for the selected node.

Sessions

Displays the current numbers of total sessions and active sessions on the selected node.

Temporary files

Displays the number of temporary files generated per second on the selected node.

Transactions

Displays the number of active transactions, the number of idle long-running transactions, the number of idle transactions, the number of XA transactions, the number of pending transactions, and the number of long-running transactions on the selected node.

Cache

Displays the block hit ratio, the number of block reads, and the number of block hits per second on the selected node.

Vacuum

Displays the difference between the current transaction ID and the earliest transaction ID that cannot be recycled.

Scanned rows

Displays the number of rows that are inserted, read, updated, and deleted per second on the selected node.

PolarDB storage usage

Displays the storage usage of WAL logs, user data, local logs, instance space, temporary space, and other storage on the selected node.

PolarDB IOPS

Displays the number of reads and writes per second for the selected node, including the total number of reads and writes per second, the number of reads per second, the number of writes per second, the total I/O throughput, the read I/O throughput, and the write I/O throughput.

Performance events

Exceptions

Displays the statistics of exceptions on a selected node.

Optimization events

Displays the statistics of optimization events on a selected node.

FAQ

  • Why does the Advanced Monitoring page display a QPS of approximately 10 queries per second despite no active service traffic on the cluster?

    The monitoring, log collection, and administration tasks that run in the system background generate approximately 10 queries per second. This has little impact on the performance of your cluster.

  • How do I resolve the high CPU utilization issue?

    We recommend that you perform the following steps:

    • Check whether a large number of slow requests exist. If a large number of slow requests exist, we recommend that you optimize slow SQL statements first.

    • Check whether the trend of the CPU utilization curve is consistent with that of the QPS or TPS curve. If the trends of the curves are consistent, the issue is caused by high transaction concurrency. In this case, we recommend that you upgrade the configuration of your cluster.

    • If no slow requests exist and the curves are inconsistent, contact technical support.

  • What do I do when the number of connections is much greater than the number of active connections?

    You can specify smaller values for the wait_timeout and interactive_timeout parameters to accelerate the release of idle connections. We recommend that you close the connections that are no longer in use in time to reduce the number of idle connections.

Related API operations

Operation

Description

DescribeDBClusterPerformance

Queries the performance data of a PolarDB cluster.

DescribeDBNodePerformance

Queries the performance data of a PolarDB node.

DescribeDBClusterMonitor

Queries the monitoring data collection frequency of a PolarDB cluster.

ModifyDBClusterMonitor

Modifies the monitoring data collection frequency of a PolarDB cluster.