Services¶
Services serve as a centralized observation entry point, presenting performance metrics, invocation topology, and full observability data (including traces, logs, infrastructure, etc.) in distributed systems from a global perspective, thus providing an integrated analysis capability from overall status overview to in-depth diagnosis of a single service. Service assets are uniformly managed in the Unified Catalog.
Metrics¶
The Metrics list provides a core dataset for measuring service health. It displays the performance data of all services in the current workspace in a list form, such as key metrics like request count, error rate, and response time.
This allows you to quickly assess the running status of each service from a global perspective, and quickly locate target services with anomalies (such as sudden increases in error rate or prolonged response time) through functional components like filtering, searching, and Time Widget, providing a clear starting point for subsequent analysis.
Service Map¶
The Service Map clearly presents the invocation relationships and overall dependency structure between services in a topology diagram. It aggregates the originally scattered trace information into a comprehensible "map", allowing you to intuitively see which upstream services a service depends on and which downstream services call it.
This global dependency view is crucial for understanding system architecture, assessing the impact scope of changes, and quickly locating the problem propagation chain when a failure occurs.
Service Assets¶
Service assets have been uniformly migrated to Unified Catalog > Service for management. You can view, add, or edit services in the Unified Catalog, and maintain related business attributes.
Through the Unified Catalog, technical entities (services) can be associated with the daily work context of R&D and operations, making it easy to quickly view related information from service assets.
Service Details¶
Click on any service to enter its service details page, which displays a comprehensive analysis interface aggregating all observability data of the service. The page seamlessly integrates different observation perspectives such as Analysis Dashboard, Resource Calls, Infrastructure Dependencies, Traces, and Logs through the top tabs.
Thus, in-depth analysis around a single service can be completed within the same interface without switching between different functional modules. For example, you can start from the macro performance chart (Analysis Dashboard), then view the most time-consuming API endpoints (Resource Calls), then trace their complete call chain (Traces), and finally check the host or container resource status that supports the service operation (Infrastructure Dependencies).
