Service Dependency Graph for Alert Flooding Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing IT systems struggle to accurately identify and maintain dependencies among services due to the complexity and non-scalability of existing IT systems, and existing systems lack the technical capabilities to automatically identify and maintain dependencies amongst services.
Innovation Solution
A relationship network graph is constructed to identify and maintain dependencies among services by representing entities as nodes and interactions as edges, allowing for automatic identification and updating of service dependencies using interaction data sources and network analysis techniques.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual declaration of service dependencies by service owners is used, then some dependencies can be recognized, but the system complexity increases and not all dependencies are identified
Solution Approach 1:
The system automatically identifies service dependencies by monitoring interactions between services without requiring manual input from service owners. The dependency graph is constructed and maintained automatically based on observed service-to-service communications, events, and alerts, enabling the system to self-determine dependencies rather than relying on human declaration.
Solution Approach 2:
The patent replaces the manual mechanical process of service owners declaring dependencies with an automated computational system that uses network graph analysis and machine learning algorithms to infer dependencies from operational data, thereby eliminating the need for manual intervention while improving accuracy.
2Loss of information
If monitoring systems send alerts to all services, then visibility into operational metrics is improved, but information flooding occurs across the system
Solution Approach 1:
The system uses the dependency graph to implement feedback-based alert routing, where alerts are selectively propagated only to services that have a documented dependency relationship with the affected service. This feedback mechanism ensures that information reaches the right recipients without overwhelming the system with unnecessary notifications.
Solution Approach 2:
The patent extracts and filters alerts based on the dependency graph, removing unnecessary alert propagations to services without dependencies. This selective extraction ensures that only relevant services receive alerts, reducing information flooding while maintaining visibility for affected services.
3Measurement precision
If automatic dependency identification is implemented, then dependency accuracy improves, but computational resources increase
Solution Approach 1:
The system implements partial monitoring by focusing computational resources on identifying and tracking only the most critical service interactions and dependencies. Rather than analyzing every possible service communication, the system uses sampling and threshold-based approaches to detect dependencies with high accuracy while consuming fewer computational resources.
Solution Approach 2:
The patent dynamically adjusts monitoring parameters such as sampling rates, alert thresholds, and graph update frequencies based on system conditions and criticality levels. This allows the system to maintain high dependency detection accuracy during critical periods while reducing computational overhead during normal operations.
Data Source
AI summary
A relationship network graph is constructed, where nodes represent entities and edges represent interactions between the entities, and wherein at least some of the entities represent respective persons, teams, or organizations of people. A dependency strength is determined from the relationship network graph between a first service associated with a first entity and a second service associated with a second entity. A first alert associated with the first service is received. In response to the first alert, and based on the dependency strength meeting a predefined threshold, a second alert associated with the second service is paused for a predefined pause period. It is determined whether the first alert remains unresolved after expiration of the pause period. If the first alert remains unresolved, the second alert is resumed.