Overlay-Underlay Network Fault Detection with Machine Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing network communication systems experience service interruptions due to faults in overlay and underlay networks, leading to inconvenient and degraded operations, and current methods rely on slow and tedious human analysis for fault detection.
Innovation Solution
A machine learning model, such as a neural network, is used to predict communication-related faults by analyzing embedding representations of network attributes, allowing for early detection and remediation actions to prevent service disruptions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If human analysis is used for fault detection, then the detection process can be performed with simple tools, but the detection speed is slow and the process is tedious
Solution Approach 1:
The patent replaces manual human analysis with an automated machine learning model that processes network attributes and predicts faults. The system uses embedding representations of network attributes as input to the ML model, which automatically generates fault predictions without requiring human intervention in the analysis process, thereby dramatically increasing detection speed while reducing the tedious nature of manual analysis
Solution Approach 2:
The system enables self-service fault detection by automatically collecting network attributes, generating embedding representations, and using the machine learning model to predict faults without human assistance. The automated pipeline handles the entire detection process from data collection to fault prediction, eliminating the need for human analysts to manually examine network communications
2Reliability
If machine learning models are used for fault prediction, then service interruption can be prevented, but the computational resources and model complexity increase
Solution Approach 1:
The machine learning model performs preliminary fault prediction by analyzing network attributes and generating predictions before actual service interruptions occur. This proactive approach allows remediation actions to be taken in advance, preventing service interruptions before they impact network operations, thereby improving reliability through anticipatory detection
Solution Approach 2:
The patent introduces embedding representations as an intermediary layer between raw network attributes and the machine learning model predictions. This embedding layer transforms complex network attributes into a standardized format that the ML model can process efficiently, managing the complexity by creating an intermediate representation that simplifies the overall system architecture
Data Source
AI summary
In some examples, a system receives a first representation of attributes associated with a network stack connected to an underlay network that couples a first system to a computing environment, where the network stack comprises a plurality of layers. The system receives a second representation of attributes associated with an overlay network provided over the underlay network. The system provides the first representation and the second representation to a machine learning model trained to detect a fault associated with communications between the first system and the computing environment. The machine learning model generates an output comprising a value representing a likelihood of a presence of the fault associated with the overlay layer or the underlay layer. Based on the output, the system initiates a remediation action to address the fault.