Federated API Subgraph Generation via Tagging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing cloud platform APIs face inefficiencies due to complex configurations and built-in latencies when accessing multiple APIs with different authentication settings, and federated APIs expose all underlying data, leading to data privacy and governance issues, as well as limitations in creating unique backend APIs for front-end applications.
Innovation Solution
Implementing subgraph generation via tagging within a federated API service, where users annotate source APIs with tags like 'confidential' or 'production', allowing selective inclusion or exclusion of elements in the subgraph, thereby generating a customized, filtered view of the federated API that limits exposure of sensitive data and enables targeted scaling.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a federated API exposes all underlying data sources to provide comprehensive access, then data accessibility is improved, but data privacy and governance issues worsen
Solution Approach 1:
The patent segments the federated API into multiple subgraphs, each representing a specific data domain or business context. This segmentation allows selective exposure of data sources through targeted subgraphs rather than exposing all data sources through a single comprehensive API, thereby maintaining data privacy while preserving accessibility for authorized access patterns.
Solution Approach 2:
The patent applies local quality by assigning different privacy and access control characteristics to different subgraphs. Each subgraph can have its own authentication settings, data source configurations, and access policies, allowing data privacy to be maintained at the local subgraph level while still providing comprehensive data accessibility across multiple specialized access points.
2Adaptability or versatility
If multiple APIs with different authentication settings are accessed to provide comprehensive service functionality, then service capability is improved, but configuration complexity and latency worsen
Solution Approach 1:
The patent merges multiple underlying APIs with different authentication settings into a unified federated API interface. The federation layer abstracts away the complexity of individual API configurations, authentication mechanisms, and data source variations, presenting a consistent interface to clients while maintaining access to diverse service capabilities through the combined subgraphs.
Solution Approach 2:
The federated API acts as an intermediary layer between clients and multiple underlying APIs. This intermediary handles authentication, data retrieval, and result aggregation, eliminating the need for clients to directly configure and manage multiple APIs with different authentication settings, thereby reducing configuration complexity while preserving comprehensive service functionality.
3Loss of information
If multiple APIs are pinged to perform a service to provide comprehensive data retrieval, then data completeness is improved, but response time and latency worsen
Solution Approach 1:
The patent implements preliminary action by pre-configuring subgraphs with their associated data sources, authentication settings, and query templates during system initialization or deployment. This preliminary setup allows the federated API to execute pre-planned data retrieval operations across multiple data sources without requiring real-time configuration or sequential authentication handshakes, thereby reducing response time while maintaining data completeness.
Data Source
AI summary
A computing system may support a federation management service that receives an input indicating a plurality of data sources comprising one or more tags corresponding to one or more elements of the plurality of data sources. The federation management service may generate a federated application programming interface (API) using the data sources, and the federated API may include elements of the data sources. The federation management service may receive a second user input selecting a set of tags for exclusion or inclusion of elements in a subgraph of the federated API. The federation management service may generate the subgraph that includes a subset of the one or more elements of the plurality of data sources.