Asynchronous Fault Domain Graph for Network Root Cause Isolation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In computer networks, identifying the root cause of faults and network issues is challenging due to the complexity of interactions among various entities, such as routers and switches, where traffic overload and network faults can lead to unreachable entities, making it difficult for network management systems to diagnose and recover from problems effectively.
Innovation Solution
A fault isolation system that uses asynchronous notifications to build a fault domain graph, requesting and updating neighbor lists, and determining the root cause by analyzing the statuses of entities within the graph, allowing for the identification of root causes even in complex network topologies with multiple issues.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a network management system performs comprehensive fault analysis by querying all entities, then the root cause can be accurately identified, but the processing time and system complexity increase significantly
Solution Approach 1:
The patent segments the network into fault domains based on connectivity relationships. Instead of analyzing all entities comprehensively, the system identifies and isolates specific fault domains containing unreachable entities, then performs targeted analysis only within these segmented regions. This reduces the analysis scope while maintaining accuracy in identifying root causes within the affected domains.
Solution Approach 2:
The patent introduces a fault domain graph as an intermediary structure that mediates between the network management system and the actual network entities. This graph represents connectivity relationships and enables the system to reason about fault propagation without directly querying all network entities, thereby reducing analysis time while preserving diagnostic accuracy.
2Reliability
If the system queries all entities and their neighbor lists synchronously, then complete fault domain information is obtained, but the processing complexity and resource consumption increase
Solution Approach 1:
The patent implements dynamic fault domain identification where the system adapts its query strategy based on the current network state and connectivity information. Instead of statically querying all entities, the system dynamically builds fault domain graphs by requesting neighbor lists only for entities that are unreachable or suspected of being faulty, reducing processing complexity while maintaining information completeness.
Solution Approach 2:
The patent performs preliminary actions by first building a fault domain graph based on connectivity information before conducting comprehensive fault analysis. This preliminary structuring organizes the data in advance, allowing subsequent fault analysis to proceed more efficiently without requiring complex synchronous queries to all entities.
3Productivity
If the system uses asynchronous notification mechanisms, then processing efficiency improves, but the coordination complexity between system components increases
Solution Approach 1:
The patent implements feedback mechanisms where network entities asynchronously notify the fault isolation system about their operational status and connectivity changes. This feedback-driven approach allows the system to update fault domain graphs in real-time without requiring complex coordination protocols, improving processing efficiency while managing coordination complexity through event-driven architecture.
Solution Approach 2:
The patent enables network entities to self-report their status and connectivity information asynchronously to the fault isolation system. This self-service mechanism eliminates the need for the system to actively query all entities, improving processing efficiency. The coordination complexity is managed by using standardized notification protocols that entities autonomously follow.
Data Source
AI summary
A fault isolation system asynchronously builds a fault domain graph to identify fault domains to isolate a root cause of a problem in a computer network. Basically, when the fault isolation system receives a notification that an entity of the network is unreachable (i.e., in a down status), the fault isolation system requests neighbor lists and statuses for the neighbors to asynchronously build a fault domain graph. From that topology, the fault isolation system determines a root cause of the problem with the network.