Phone:400-882-3320
Guance 
Customer context
Wangxiaobao provides AIoT and data services for sales-intelligence scenarios. Its published story describes monitoring expanding from individual systems to multi-cloud resources and core business workflows.
Existing open-source tools addressed specific monitoring needs, but environments and business paths retained separate data and alert entry points.
Across complex scenarios, the team needed one way to understand system health, fault boundaries, cross-cloud dependencies, and alert context.
Local monitors could not explain complex paths
Multi-cloud resources, microservices, and business workflows lacked a shared object model, forcing investigators to change platforms.
Fault isolation depended on manual experience
Information remained separate across teams, leaving major events without a shared evidence set and timeline.
Repeated alerts consumed operator attention
Several tools produced duplicate or low-value notifications that had to be sorted before useful investigation began.
Implementation
Connect full-stack data with DataKit and shared tags
DataKit collected runtime data from different environments while common tags related metrics, logs, traces, and resources by service and environment.
Represent resource relationships in a cross-cloud topology
Compute resources and dependencies across clouds entered a shared topology for drill-down from health state to related objects and monitoring configuration.
Correlate traces, logs, and metrics for microservices
Call topology reconstructed request paths and placed related logs, metrics, and alerts in the same time context to help identify a fault boundary.
What changed
System health gained a shared view
Operators could review anomalies by service, cloud environment, and resource relationship without maintaining separate incident narratives.
Alerts could be grouped around objects
Shared tags and topology provided a basis for grouping repeated signals by service and resource.
Multi-cloud dependencies entered one investigation
Cross-cloud resource state could be reviewed together with microservice traces, logs, and metrics.
Platform work shifted towards data quality
After consolidating separate entry points, teams could focus on tag standards, alert policies, and reliability workflows.
