Performance Metrics¶

On the Performance Metrics page, you can switch between the Service and Endpoint views to view the request volume, response time, and errors of services or endpoints, helping you quickly locate performance issues. The Service view supports quick filtering by service type, environment, version, project, and service name.

Service / Endpoint Views¶

• Service: View performance metrics by service, including total requests, average request count, average response time, P75 response time, P99 response time, maximum response time, error count, and Apdex score.
• Endpoint: Aggregate by service and HTTP route (http_route) to view the request volume, response time, and errors of each endpoint. Supports searching by route and filtering by service.

For example, the /users/{id} route of the same service is aggregated into one Endpoint, while routes with the same name in different services are counted separately.

Workspace Scope¶

By default, data from the current workspace is shown. Using the "Workspace" selector at the top, you can select multiple authorized workspaces within the same site to query service performance jointly. For authorization requirements and read-only restrictions, see APM and RUM Cross-Workspace Queries in the Same Site.

Services with the same name in different workspaces are displayed separately, and the list indicates the source workspace of each. Hover over a workspace name to view its UUID. Search, sorting, pagination, and count statistics are all based on the currently selected workspace scope.

After clicking a service, the details open in the workspace to which that record belongs; returning to the list preserves the original query scope. In cross-workspace read-only mode, you cannot add services, toggle favorites, or modify monitor configurations.

Add Services¶

Manual Installation

Auto-Injection

Manage Performance Metrics¶

You can manage performance metrics through the following actions:

• Use the Time Widget to select a time range and view the corresponding performance data;

• In the Service view, choose the columns to display in the list, including Workspace, Service, Language, Total Requests, Average Request Count, Average Response Time, P75 Response Time, P99 Response Time, Maximum Response Time, Error Count, Apdex Score, Team, and Monitor;

• Export the current query results as a CSV file; cross-workspace results include the workspace name and workspace UUID;

• In the Service view, hover over a single service:

  • Click to jump to the trace detail page; clicking a service directly opens the analysis dashboard for that service;

  • If a monitor has tags for the current service, click to jump to the corresponding monitor configuration page;

  • View the event status; clicking the status opens the Event Explorer page in a new tab.

Configure Filter Fields¶

In addition to the default filter fields (service, env, version, project, name), you can customize up to 5 fields as service filter fields (❗️Fields added here only take effect on new data).

1. Go to the Performance Metrics page;
2. Click the "Settings" button;
3. Add fields;
4. Save.

Data Source Definition and DQL Queries¶