Network Management Test Intent Synthesis via Machine Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Intent-based networking systems require thorough testing to ensure that intent fulfillment systems can effectively implement network changes without explicit configuration changes, as human operators need to verify the system's ability to achieve desired outcomes, but manual testing is slow, error-prone, and may not cover all possible network modifications.
Innovation Solution
A machine learning mechanism or rule-based synthesizer generates a set of test instructions that trigger network changes without configuring individual nodes, using a block of text and input instructions to produce diverse and valid test intents, allowing for the evaluation of network responses and corner scenarios.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If manual testing methods are used to verify intent fulfillment systems, then testing can be performed with simple procedures, but the testing process is slow and error-prone
Solution Approach 1:
The patent replaces manual mechanical testing processes with an automated system that uses natural language processing and machine learning models to generate and execute test cases. The system automatically parses intent descriptions, generates corresponding test instructions, and validates network configurations without human intervention, thereby increasing both speed and reliability of testing.
Solution Approach 2:
The testing system is designed to be self-service by automatically generating test cases from intent descriptions without requiring manual creation. The system autonomously parses natural language intents, generates appropriate test instructions, executes them against the network, and validates results, eliminating the need for manual test case development and execution.
2Reliability
If comprehensive testing of all network modifications is performed, then system reliability is improved, but the complexity and time required for testing increases
Solution Approach 1:
The patent implements a universal testing framework that can handle multiple types of network modifications and intent descriptions through a single system. The natural language processing component and template-based generation mechanism allow the system to accommodate various network configurations and modification scenarios without requiring separate testing systems for each case.
Solution Approach 2:
The system manages complexity by dynamically adjusting testing parameters based on the specific intent description and network configuration. The machine learning model adapts test generation parameters such as test case depth, modification scope, and validation criteria according to the characteristics of each intent, allowing comprehensive testing while maintaining manageable system complexity.
3Manufacturing precision
If detailed configuration changes are specified in test instructions, then precise control over network modifications is achieved, but the flexibility of intent-based networking is reduced
Solution Approach 1:
The patent introduces an intermediary layer consisting of natural language processing and template-based generation mechanisms that translate high-level intent descriptions into detailed configuration instructions. This intermediary preserves intent flexibility by allowing operators to specify goals in natural language while automatically generating the precise configuration changes needed, without requiring operators to manually specify detailed technical parameters.
Solution Approach 2:
The system segments the test instruction generation process into distinct components: intent parsing, test case generation, configuration modification, and validation. This segmentation allows the system to maintain flexibility at the intent level while achieving precision at the configuration level, with each segment handling a specific aspect of the transformation from high-level intent to detailed configuration changes.
Data Source
Figure 1
Figure 2A~2B
Figure 2C~2E
AI summary
According to an example aspect of the present invention, there is provided an apparatus comprising at least one processing core, at least one memory including computer program code, the at least one memory and the computer program code being configured to, with the at least one processing core, cause the apparatus at least to generate, using a machine learning mechanism, a set of test instructions, wherein each one of the test instructions is configured to trigger a first network to change one or more technical characteristics of the first network, the test instructions not comprising configuration changes in nodes of the first network that are usable in changing the one or more technical characteristics of the first network, and use a block of text as input in the generation of the set of test instructions.