DCF Node Authentication Using Application Whitelists
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data confidence fabrics (DCF) lack a mechanism to identify unexpected data sources and filter out non-required traffic, as nodes cannot determine if data received is from an unexpected source or not.
Innovation Solution
Implementing a device onboarding process with application discovery to whitelist trusted applications, enabling DCF nodes to recognize and manage data from known sources, and apply confidence scores or take actions like rejection or annotation for unknown data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If DCF nodes process all received data without authentication, then data processing throughput is maintained, but data integrity and trustworthiness deteriorate due to inability to identify unexpected data sources
Solution Approach 1:
The system performs preliminary authentication by maintaining a whitelist of authorized data sources at DCF nodes. Before processing data, the node checks whether the data source is in the whitelist, thereby preventing untrusted data from entering the processing pipeline. This preliminary filtering action resolves the contradiction by ensuring data trustworthiness without requiring complex real-time analysis of every data packet.
Solution Approach 2:
The patent introduces an intermediary authentication layer between data sources and DCF nodes. This intermediary mechanism uses pre-configured whitelist information to mediate the trust relationship, allowing nodes to authenticate data sources without direct complex verification. The intermediary whitelist acts as a trusted third party that resolves the contradiction by simplifying the authentication process while maintaining reliability.
2Reliability
If DCF nodes implement authentication mechanisms to filter unexpected data sources, then data integrity improves, but processing speed and productivity deteriorate due to additional verification steps
Solution Approach 1:
Authentication credentials and whitelist information are pre-configured in the system before runtime data processing. This preliminary setup allows DCF nodes to perform simple lookup operations during data processing rather than executing complex verification algorithms, thereby maintaining high processing throughput while ensuring data integrity through pre-established trust relationships.
Solution Approach 2:
The system uses a copied whitelist data structure that is replicated across DCF nodes, allowing each node to independently authenticate data sources using local copies of the authorization information. This copying approach eliminates the need for real-time communication with a central authentication server, thereby maintaining processing speed while ensuring data integrity through distributed authentication.
3Adaptability or versatility
If DCF nodes accept data from any source, then system adaptability and ease of operation are maintained, but security and control deteriorate due to inability to filter non-required traffic
Solution Approach 1:
The patent implements local quality control by allowing different DCF nodes to have different whitelists tailored to their specific requirements and trusted data sources. Each node can customize its authentication criteria locally while maintaining overall system adaptability. This local quality approach resolves the contradiction by enabling nodes to filter harmful untrusted data while preserving system flexibility through node-specific configurations.
Data Source
AI summary
One example method includes receiving, at a node of a data confidence fabric, data from a data source, determining, by the node, whether the data source is listed in a portfolio of data sources, when the data source is listed in the portfolio, processing, by the node, the data, when the data source is not listed in the portfolio, performing, by the node, an analysis of the data, when the analysis reveals that the data is recognized by the node, processing, by the node, the data, and when the analysis reveals that the data is not recognized by the node, performing an action.