Entity topology in Cloud Monitor 2.0 provides a layered, interactive view of entity dependencies and metrics. Use it to identify bottlenecks, spot performance hotspots, and trace issue propagation paths.
Procedure
-
Log on to the Cloud Monitor 2.0 console and navigate to any workspace.
-
Open any entity details page and choose .
Tools
-
Toolbar (upper-left corner)
Supports four operations: Focus, Center, Reset, and Filter.
-
Minimap (lower-right corner)
Drag the minimap to pan across the topology.
-
Legend (lower-left corner)
Shows all entity types in the topology. Click a type, such as Interface, to highlight its nodes with a breathing animation and its edges with a flow animation.
By default, shows two rows. Click the expand icon to show all types.
-
Filter pane
Filters the topology by Node Type, Edge Type, and full-text search. Selections within the same category use OR logic; different categories use AND logic.
For example, selecting the instance and application node types and entering gateway in full-text search displays all matching entities.
Full-text search matches all key-value pairs in an entity.
Node interaction
Node types
The topology displays two node types:
-
Single node: Shows the entity name and key metrics, wrapping up to two lines. Supports five interactions: Collapse, focus, View Metrics, expand next level, and clicking to open the information pane.
-
Group node: Shows the type name and entity count. Supports three interactions: Expand, View Metrics, and split child nodes.
Clicking a node
-
Clicking a single node opens the Entity Information pane, which shows entity details and metric charts for the selected time range.
-
Clicking a group node opens the split child nodes pane. Select entities to detach from the group and display them individually.
Expanding and collapsing
Hover over a group node and click Expand Group to display all nodes individually.
Click Collapse Group to collapse expanded groups.
Groups with more than 8 entities cannot be expanded directly. Click the group node to open the split child nodes pane and select specific nodes to display.
If an expand next level action generates new nodes that cause the total number of nodes in an expanded group to exceed 8, the previously expanded child nodes are added to the split list, and the new nodes are placed into a group node.
Focus
Hover over a node to enter focus tracking mode. The view centers on the focused node and shows its directly connected upstream and downstream nodes. Each expand next level action shifts the focus to the newly expanded node.
In focus mode, group nodes contain only nodes related to the focused node. Expanding a group and clicking Cancel Focus moves those nodes to the split child nodes pane.
Metrics
-
For a single node, hover over it and click View Metrics to open a pane showing all metrics. The content matches the Metric Explorer page.
-
For a group node, hover over it and click View Metrics to select and compare metrics side-by-side across entities in the group.
Expand next level
Hover over the expand next level icon on the right side of a single node to explore its directly connected upstream and downstream entities.
If an expand next level action causes the current node to be absorbed into a group node, the system automatically adds the original node to the split child nodes list to ensure it remains available as an interaction target.
Edge interaction
Clicking an edge
If an edge has metrics, the topology displays up to two recent data points on it. Click the edge to open the Edge Metric Information pane.
Intelligent layout
The topology uses a proprietary, multi-level subgraph layout algorithm.
-
Hierarchical layout: Nodes are layered vertically by domain, from top to bottom: Real User Monitoring (RUM), Application Monitoring, container orchestration (such as Kubernetes), Alibaba Cloud services, and infrastructure. Within each domain, nodes are grouped by type and arranged by the layout algorithm.
-
Dependency-driven directed layout: Within the same domain, nodes are arranged horizontally by dependency so that edges flow left to right, except for cyclic dependencies.
-
Sub-topology optimization: An expanded group of same-type nodes with no internal connections occupies its own column to prevent overlap. If internal connections exist, a dedicated sub-layout is applied.
-
Cycle handling: Cyclic relationships use a special layout. Levels are assigned based on node out-degree to ensure readability.