View the performance of your database fleet

Preview — View database fleet performance

This feature is subject to the "Pre-GA Offerings Terms" in the General Service Terms section of the Service Specific Terms. You can process personal data for this feature as outlined in the Cloud Data Processing Addendum, subject to the obligations and restrictions described in the agreement under which you access Google Cloud. Pre-GA features are available "as is" and might have limited support. For more information, see the launch stage descriptions.

Database Center provides performance information about your fleet resources to help you maintain your database fleet. You can view performance insights that help you identify specific performance issues, such as latency due to CPU saturation. You can also view a list of your database resources to identify the most highly utilized resources for different metrics, such as CPU utilization and storage used. In addition, you can monitor active queries across your fleet to identify and analyze query issues, such as slow queries. Use this information to prioritize database fleet maintenance and ensure optimal fleet operation.

Before you begin

• Set up Database Center for your organization.

• To enable the Gemini Cloud Assist button, ensure that you have the geminicloudassist.investigations.create Identity and Access Management (IAM) permission. For more information, see Grant or revoke a single IAM role.

• To view fleet performance insights, esure you have the roles/recommender.viewer (iam_name_short) permission.

View database fleet performance

To view your database fleet performance details, go to the Performance page in Database Center.

Go to Database Center

The Performance page includes a Fleet performance insights section and two tabbed sections, Performance metrics and Queries.

Fleet performance insights

The Fleet performance insights section displays cards that highlight performance insights that are generated by Gemini. Use these insights to help you identify and understand specific issues in your database fleet.

Each card includes a View resources link. When you click View resources, the following information appears:

• A detailed description of the issue.
• A table listing the resources that exhibit the issue. The table includes the database product, the Google Cloud project, and metrics specific to the issue.

From the table, click the name of a resource to view its details in the product that uses it.

View performance metrics

The Performance metrics tab includes one card for each database product in your fleet. The cards appear in alphabetical order by product name. A card appears for each database product you use, such as Cloud SQL and Spanner. Product cards don't appear for products you're not using. Each product card includes the following performance details:

• Resource count: The total number of resources used by the database product.

• Metric selection: Select a metric from the available metrics. The metrics you can choose from are pertinent to the database product. For example, the peak storage utilization metric is available on a Cloud SQL card, but not the AlloyDB for PostgreSQL card.

• Resources ranked by utilization: The table on the product card lists resources in the Resource name column by utilization for the selected metric. The table lists the resource with the highest utilization first. Click the resource name to view details about that resource. The table shows ten resources. Click or to show the previous or next ten resources.

• Resource utilization over time: For each resource listed in the table, the chart shows resource utilization over the past seven days for the selected metric. Hold the pointer over a line to see which resource it's associated with.

Monitor active queries

The Queries tab lets you monitor active queries across your fleet to identify and analyze query issues, such as slow queries. This feature supports the AlloyDB and Cloud SQL database products. It also displays a list of top long-running transactions with normalized active queries. Each query represents a group of resources that a database product uses. For example, one row might show a query for a group of resources that AlloyDB uses.

A normalized active query removes sensitive data and returns a digest. The digest is the same for different values. For example, select * from test_table where id=1; and select * from test_table where id=2; both have the digest select * from test_table where id=?;.

The Queries tab contains a table with detailed information about each normalized query, such as the following:

• The number of resources the query uses. Click the link to view resource details.

• The database product and engine to which the resources belong.

• The Google Cloud project that uses the resources. If you're using an organization-based hierarchy in Google Cloud, then a query can cross Google Cloud project boundaries.

• The average number of rows the query returns.

• The total and average CPU time for the query. By default, queries are sorted by the number of times they are called, so the most frequently used queries appear at the top of the list.

• The number of times the query ran.

View query resource details

When you click a link in the resources column, a table appears with detailed information for each resource, such as the following:

• The name of the resource. Click the name to view the query insights page for that resource.

• The database product and engine to which the resource belongs.

• The average and total amount of time it took for the query to run on the resource.

• How many times the query ran on the resource.

• The average number of rows the query returns on the resource.

What's next

• For more information about Database Center, see the Database Center overview.

• For information about analyzing your database inventory, see Analyze your database inventory.

• For information about analyzing database fleet health, see Analyze database fleet health using Gemini.