Tokenized Key-Value Data Storage for Non-Sensitive Database Security
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing database systems store sensitive information that is prone to attacks, as they handle and store sensitive data in a hierarchical structure, lacking security and integrity, especially in cases where attackers gain access to security keys or references.
Innovation Solution
A system that receives data in a non-sensitive format, splits it into chunks, generates by-value and by-reference tokens, and stores these in a key-value table, ensuring data remains non-sensitive until the by-reference token is used for extraction, utilizing split-token encryption and NoSQL databases for enhanced security.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If data is stored in a traditional database format, then data accessibility and operational simplicity are improved, but data security and vulnerability to attacks deteriorate
Solution Approach 1:
The patent divides data into multiple separate silos or databases, where each silo contains only a portion of the complete data set. This segmentation ensures that no single database contains all sensitive information, reducing the impact of potential breaches while maintaining operational accessibility through coordinated queries across multiple silos.
Solution Approach 2:
The patent introduces an intermediary component that manages data retrieval across multiple silos. This intermediary coordinates access to distributed data portions, allowing seamless data accessibility for authorized operations while preventing direct access to individual silos, thus maintaining both ease of operation and enhanced security.
2Reliability
If data is encrypted and split into multiple silos, then data security is improved, but system complexity and data retrieval difficulty worsen
Solution Approach 1:
The patent implements a universal data access layer that handles multiple functions including data retrieval, encryption management, and coordination across different silos. This multi-functional approach consolidates complexity into a single management interface, maintaining security through distributed storage while simplifying operations through a unified access mechanism.
3Speed
If all data is stored in a single database, then data retrieval speed is improved, but vulnerability to unauthorized access and data breaches worsens
Solution Approach 1:
The patent segments data across multiple databases or silos, so that no single database contains the complete data set. This segmentation reduces the attack surface for unauthorized access while maintaining retrieval speed through optimized distributed query mechanisms that efficiently aggregate data from multiple sources.
4Reliability
If data is devalued and tokenized, then security against attackers is improved, but data usability and processing capability deteriorate
Solution Approach 1:
The patent creates tokenized copies of data that can be used in place of sensitive information. These tokens maintain the necessary data structure and relationships for processing and analysis while being useless to attackers without the ability to map them back to original data, thus preserving data usability while enhancing security.
Data Source
AI summary
A method for storing data in a database in a devalued format is provided. In one aspect, a system receives at an API endpoint a chunk of data sent from a client computer. The system generates a by-value token and a by-reference token in response to the chunk of data received at the API endpoint. The system transmits the by-reference token to the client computer. The system indexes the chunk of data in the database in a key-value table, where a first column and second column of the key-value table comprises the by-value token and a chunk of data respectively. The data is devalued such that it remains non-sensitive until the client computer sends the by-reference token which can be used with the by-value token to extract the chunk of data from the database.