API Registry Knowledge Graph Change Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Managing software API updates is challenging due to asynchronous communication between providers and customers, unknown impact of API changes, and the difficulty in identifying breaking changes, leading to potential crashes and resource wastage, especially in distributed and cloud-based environments.
Innovation Solution
An API registry system that transforms API software code into knowledge graphs, identifies changes between versions, and sends customized alerts to subscribers, allowing for simulation of API upgrades and downgrades to assess impact and inform decision-making.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If traditional monolithic architecture is used for software development, then API management is simpler, but adaptability to distributed and cloud-based environments is reduced
Solution Approach 1:
The patent segments the monolithic software architecture into microservices, allowing each service to have its own API lifecycle. This segmentation enables distributed deployment and independent scaling while the central registry coordinates across segments to manage overall API governance.
Solution Approach 2:
The patent introduces an API registry as an intermediary component that mediates between microservice providers and consumers. The registry receives API definitions, generates knowledge graphs, and distributes update notifications, thereby simplifying the complexity of managing distributed APIs without requiring direct peer-to-peer communication between all services.
2Loss of information
If asynchronous communication is used between API providers and customers, then system independence is maintained, but information synchronization is reduced
Solution Approach 1:
The patent implements a feedback mechanism where the API registry continuously monitors API definitions and automatically generates update notifications when changes are detected. Subscribers receive synchronous notifications about API updates, ensuring information synchronization while maintaining the independence of provider and consumer systems through the registry intermediary.
3Measurement precision
If comprehensive API monitoring is implemented to identify all consumers, then impact analysis accuracy is improved, but system complexity increases
Solution Approach 1:
The API registry acts as a central intermediary that automatically tracks all API consumers and their subscription relationships. This centralized monitoring approach provides comprehensive impact analysis capability without requiring complex distributed tracking mechanisms at each service endpoint, thereby improving measurement precision while controlling system complexity.
4Productivity
If manual API update management is performed, then resource allocation is controlled, but productivity is reduced
Solution Approach 1:
The patent implements self-service automation where the API registry automatically generates knowledge graphs from API definitions, identifies changes between versions, determines impacted consumers, and distributes update notifications. This automation eliminates manual impact assessment processes, significantly improving productivity while reducing the time loss associated with manual analysis and notification distribution.
Data Source
AI summary
A computer-implemented method can include registering a new version of an application programming interface (API) with an API registry, transforming the new version of the API to a first knowledge graph, comparing the first knowledge graph to a second knowledge graph to determine a difference graph, and sending the difference graph to selected entities who have subscribed to the API registry. The second knowledge graph can be transformed from a prior version of the application programming interface. The difference graph connects the second knowledge graph to the first knowledge graph and identifies changes from the second knowledge graph to the first knowledge graph.


