API Caching Layer Segments Personal Data for Privacy

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing API systems face inefficiencies when handling requests that mix public, personal, and customized information, leading to unnecessary duplication and processing overhead due to privacy controls, as responses containing personal data cannot be cached, affecting performance and bandwidth usage.

Innovation Solution

Implementing an API caching and orchestration layer that separates personal, customized, and public information, allowing each portion to be cached or handled appropriately, with the orchestration layer routing requests to specific microservices and managing responses to maintain privacy while enabling caching of non-personal data.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If responses containing personal information are subject to privacy controls and cannot be cached, then privacy protection is improved, but API performance and system efficiency deteriorate due to unnecessary duplication and processing overhead

Engineering Contradiction:
Improveprivacy protectionVSAvoidAPI performance
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent segments the API response into multiple parts: personalized information (containing personal data) and non-personalized information (public data). Each segment is handled differently - personalized segments are not cached to protect privacy, while non-personalized segments are cached for performance optimization. This segmentation allows the system to simultaneously achieve privacy protection and maintain API performance by caching only the appropriate portions.

Inventive Principle:
Principle #1Segmentation

2Reliability

If all requests containing personal information are processed without caching, then privacy protection is maintained, but bandwidth usage and processing time increase unnecessarily

Engineering Contradiction:
Improveprivacy protectionVSAvoidbandwidth usage
Core Design Contradiction:
ReliabilityVSLoss of energy

Solution Approach 1:

The response is segmented into personalized and non-personalized portions. The non-personalized portion (such as product catalogs, public information) is cached and served from cache for repeated requests, significantly reducing bandwidth usage. Only the personalized portion is excluded from caching, ensuring privacy protection while minimizing energy loss through selective caching of appropriate content.

Inventive Principle:
Principle #1Segmentation

3Reliability

If personalized information taints the entire response as customized, then privacy protection is ensured, but caching capability is lost for the entire response including public data

Engineering Contradiction:
Improveprivacy protectionVSAvoidcaching mechanism complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

Instead of treating the entire response as customized when it contains any personal information, the patent segments the response into distinct personalized and non-personalized parts. This allows the caching mechanism to operate on non-personalized segments while excluding only the personalized segments from caching. The approach maintains privacy protection without unnecessarily complicating the caching mechanism, as the segmentation logic is applied at the data level rather than requiring complex cache management rules.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP4143720B1Efficient API with privacy protection
Publication Date: 2024.11.20 SALESFORCE INC
  • EP4143720B1 patent drawingFigure 1
  • EP4143720B1 patent drawingFigure 2
  • EP4143720B1 patent drawingFigure 3A~3B

AI summary

Methods, computer readable media, and devices for efficient application programming interface (API) processing with privacy protection are disclosed herein. One method may include receiving a user request for content from a client, parsing the user request for content to identify one or more request portions having a type of public information, customized information, and personal information, transmitting the one or more request portions to a corresponding microservice based on the type of the portion of content being requested, receiving one or more response portions, determining a type of the one or more portions of content, and in response to determining the type of a portion of content is not personal information, caching the portion of content based on the type of the portion of content, combining the one or more response portions into a user response, and sending the user response to the client.