Encrypted Database Key Table Sync in WORM Read-Write Nodes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In fully encrypted databases, the challenge lies in efficiently managing and synchronizing data encryption key information across distributed nodes without relying on external encryption key management systems, which introduces security risks and operational inefficiencies.
Innovation Solution
A write once read many (WORM) architecture is implemented, where encryption key table updates are managed by a write node and synchronized to read nodes, using internal database functions and trusted execution environments to ensure secure and timely key updates without external dependencies.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If external encryption key management systems are used, then encryption key information can be stored and managed, but security risks increase and operational efficiency decreases
Solution Approach 1:
The patent extracts the encryption key management function from external systems and integrates it into the database itself. The encryption key table is stored within the database, and the database engine directly manages key updates and synchronization, eliminating dependencies on external key management systems and reducing architectural complexity while maintaining security.
Solution Approach 2:
The database system performs self-service by automatically managing encryption key information. When data is written or modified, the database automatically updates the encryption key table and propagates changes to read nodes without requiring external intervention. This self-managing capability reduces operational complexity and improves efficiency.
2Reliability
If encryption key table updates are performed on read nodes, then data synchronization can be achieved, but write operations on encrypted data become complex
Solution Approach 1:
The patent segments database nodes into write nodes and read nodes with distinct responsibilities. Write nodes handle all write operations and encryption key table updates, while read nodes only perform reading operations and receive synchronized key information. This segmentation simplifies write operations by confining them to dedicated write nodes, while read nodes automatically receive updated key information through synchronization mechanisms.
Solution Approach 2:
The patent introduces an intermediary synchronization mechanism that automatically propagates encryption key table updates from write nodes to read nodes. This intermediary process handles the complexity of maintaining consistent encryption key information across all nodes without requiring manual intervention or complicating the write operation interface for users.
3Reliability
If data encryption key information is stored in the database, then full link encryption is achieved, but key management operations become more complex
Solution Approach 1:
The patent makes the database engine universal by enabling it to perform multiple functions: storing encrypted data, managing encryption key information, updating key tables, and synchronizing changes across nodes. This multi-functionality consolidates what would otherwise require separate key management systems into a single database platform, reducing overall system complexity while maintaining full link encryption security.
Data Source
AI summary
A database operating method and a fully encrypted database are disclosed. The method is executed by a read node in a write once read many database and includes: receiving a reading operation request from a user, determining that a writing operation is required for a data encryption key table in response to the reading operation request; establishing a connection with a write node and forwarding the writing operation; acquiring the data encryption key table that the write node has updated based on the writing operation; generating, based on the updated data encryption key table, a reading operation response and sending the reading operation response to the user. Thus, the present disclosure proposes a write once read many solution applicable to a fully encrypted database, which solves the problem of writing encryption key information of the read node in a distributed database scenario.