Service Map¶

On the Service Map page, you can intuitively view the call relationships between services via a topology diagram.

Workspace Scope¶

When you enter the Service Map page, the system displays only the service nodes and call relationships within the current workspace by default. If cross-workspace authorization is not configured, service nodes, call edges, and metric data from other workspaces are invisible.

If the current workspace has been authorized to access data from other workspaces, you can switch the viewing dimension using the Workspace Scope selector at the top of the page:

Current Workspace: Displays the service map of the current workspace only (default).
All Workspaces: Displays the service map of the current workspace and all accessible workspaces.
Authorized Workspaces: Select multiple authorized workspaces to view their service maps.

Search, filtering, service count statistics, node colors, and topology metrics on the page are computed in real time based on the currently selected workspace scope. If the current workspace has no cross-workspace authorization, the page does not display options for other workspaces.

Note

The default latency for service map data is 5 minutes. You can adjust the max_delay parameter between 1 and 5 minutes (minute granularity) (❗️Deployment Plan only) to meet your business real-time requirements.

For SaaS Commercial Plan, the data visibility scope is configured via cross-workspace authorization. Only workspace owners, administrators, or users with the cross-workspace authorization management role can configure it. After authorization, the authorized workspace can view the corresponding data in ServiceMap; after authorization is revoked, the workspace can no longer view the source workspace's ServiceMap data.
For Deployment Plan, the data visibility scope is configured at the site level.

Single Service¶

Hover over a single service node; the corresponding service is highlighted, and other unrelated services and connection lines fade. You can view key metrics such as request count, error rate, average response time, P99/P95 response time, and maximum response time for that service.

View Service Details¶

By default, the system intuitively displays the call relationships between services within the selected workspace scope. Click any service node to slide out a panel showing detailed information about that service. It mainly includes the following elements:

Service name and its corresponding team, contact information, alert count, etc.;
Basic metrics for the current service: request count, error rate, average response time, P99/P95 response time, and maximum response time;
Click to view the current service's upstream and downstream dependencies, service overview, and associated logs and traces;
Anomaly trend charts for metrics (service request response time, error request distribution) and logs (error log count).

Manage Topology¶

Zoom in, zoom out, or fit the entire service map to the view;
Select different performance metrics to match service node colors, including request count, P50/P75/P95/P99 response time, maximum response time, and error rate;
Custom colors: Set color range intervals for the selected Fill Metric on service nodes.
Custom intervals: Define custom color range intervals for the selected Fill Metric on service nodes. The system automatically divides the topology into 5 numerical intervals based on the metric's maximum and minimum values, each corresponding to a color;
- Node size: Nodes have 3 sizes. The Max and Min values of the service's requests per second are divided into three equal ranges, each corresponding to one of the three sizes.

Note

When the requests per second is less than 0.01, the displayed value is uniformly labeled as < 0.01. However, the node size calculation is still based on the actual value.

Topology configuration:
- Distinguish by environment and version: Draw the topology graph based on two dimensions: service and service+env+version. When enabled, the system draws the service map according to different environment versions.
- Service grouping: Group topology nodes by Kubernetes cluster, environment, or service tags. Different groups are shown in separate areas, with cross-group calls still represented by connection lines.
- Distinguish same-name services by project: When enabled, the system splits services with the same name based on the project attribute. Data with the same service name but different project values are displayed as different nodes; data without a project is grouped under "Unset Project".
- Calculate node size by request count (toggle).
- Highlight scope on hover: Directly related nodes (upstream/downstream) / Full-chain related nodes (multi-level upstream/downstream).
- Show only services matching filter criteria: When enabled, the service map filters service nodes according to the current filter criteria, keeping only the nodes that meet the conditions and hiding those that do not.

Cross-Workspace Query¶

Prerequisites

All Commercial Plan workspaces belong to the same Organization ID, i.e., the same organization.

If the same Trace data is split across workspaces, the data belongs to different workspaces, and you cannot view the complete trace data in one workspace. Cross-workspace Service Map queries ensure continuity of data queries. You can directly click the button in the current workspace to view the upstream and downstream call topology of the current service.

Data Source Definition and DQL Query¶

For more details, refer to Service Map Data Source Definition.