Targeted Probing for Silent Failure Pinpointing in Large Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Identifying silent failures in large-scale computer networks is challenging due to their unannounced nature and the resource-intensive process of active monitoring, which can be demanding in terms of time and resources.
Innovation Solution
A system and method that utilize a network monitoring agent on each host node to maintain route-data in a database, identifying a subject intermediary node for investigation by selecting target probe paths and testing them with targeted probes to determine the operational status of intermediary nodes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If active monitoring is performed to identify silent failures, then fault detection capability is improved, but time and resource consumption increases
Solution Approach 1:
The patent segments the network into multiple zones and divides monitoring responsibilities among zone controllers and probing controllers. This segmentation allows distributed fault detection without requiring centralized monitoring of the entire network, reducing time and resource consumption while maintaining comprehensive fault detection capability.
Solution Approach 2:
The patent performs preliminary actions by pre-configuring route data in databases, pre-identifying potential faulty intermediary nodes, and pre-establishing probe paths. This preliminary preparation enables rapid fault detection when failures occur, improving response time without requiring extensive real-time resources.
2Measurement precision
If comprehensive network monitoring is implemented, then fault detection accuracy is improved, but system complexity increases
Solution Approach 1:
The patent introduces intermediary components including zone controllers, probing controllers, and network monitoring agents that mediate between host nodes and the fault detection system. These intermediaries simplify the overall system architecture by distributing functions and providing standardized interfaces, reducing system complexity while maintaining accurate fault detection through coordinated monitoring.
3Measurement precision
If targeted probing is performed on multiple probe paths, then fault identification precision is improved, but resource demands increase
Solution Approach 1:
The patent applies local quality by focusing probing resources on specific target intermediary nodes identified as potential fault sources, rather than uniformly monitoring all network nodes. The system selectively directs probe paths through suspected faulty nodes based on route data analysis, improving fault identification precision while concentrating resources on high-probability areas.
Solution Approach 2:
The patent employs partial action by selecting a subset of probe paths for targeted probing rather than exhaustively testing all possible paths. The system identifies and probes only the most relevant paths through suspected faulty intermediary nodes, achieving sufficient fault identification precision without the resource expenditure required for complete path coverage.
Data Source
AI summary
Systems and methods for locating network errors. The system includes a plurality of host nodes in a network of host nodes and intermediary nodes, and a database storing route data for each of a plurality of host node pairs. The system includes a controller configured to identify a subject intermediary node to investigate for network errors and select, using route data stored in the database, a set of target probe paths. Each target probe path includes a respective pair of host nodes separated by a network path including at least one target intermediary node, which is either the subject intermediary node or an intermediary node that is a next-hop neighbor of the subject intermediary node. The controller is configured to test each target probe path in the set of target probe paths and to determine, based on a result of the testing, an operational status of the subject intermediary node.