Content-Filtered Database Replication for Transaction Consistency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing database replication technologies face challenges in maintaining transaction consistency and security, particularly when data needs to be replicated across networks with different security levels or access restrictions, leading to potential data leakage and integrity issues.
Innovation Solution
A unidirectional gateway that modifies and annotates database transaction files in a human-readable format, using XML, to ensure transactional consistency and security by filtering and cleansing sensitive data, while maintaining a configurable degree of consistency through granular detection and metadata insertion.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If change data capture system reads transaction log files directly from source database server, then data replication can be performed efficiently, but the system cannot access transaction log files when source database is on different network or in restricted environment
Solution Approach 1:
The patent introduces a content filter as an intermediary component between the source database and the change data capture system. The content filter receives transaction log files from the source database server and makes them available to the capture system through a standardized interface, enabling the capture system to access transaction logs regardless of the source database's network location or security restrictions.
2Reliability
If content filter is applied to security data during replication, then sensitive information can be protected, but transaction consistency may be compromised due to data modification or removal
Solution Approach 1:
The patent applies preliminary action by having the content filter process and annotate transaction data before it reaches the change data capture system. The filter modifies sensitive data fields and adds annotations indicating the type and extent of modifications made, so that the capture system can understand the original transaction state and maintain consistency despite the filtering.
Solution Approach 2:
The content filter provides feedback to the change data capture system through annotations in the transaction log files. These annotations communicate the filtering actions taken, enabling the capture system to adjust its replication behavior accordingly and maintain accurate transaction consistency records.
3Productivity
If state of art database replication is used across networks with different security levels, then data can be replicated between source and target databases, but sensitive information may leak from source to target or from target to source
Solution Approach 1:
The content filter serves as a security intermediary positioned between databases with different security levels. It inspects, modifies, and annotates transaction data before replication, preventing sensitive information from leaking while maintaining the replication capability. The filter can be configured to block specific data types or fields based on security policies.
4Reliability
If content filter modifies transaction data, then sensitive information can be cleansed, but the database may become inconsistent if ingestion is unaware that filtration occurred
Solution Approach 1:
The content filter performs preliminary modification of transaction data and simultaneously adds annotations that document the filtering actions. This preliminary action ensures that the change data capture system receives both the modified data and the metadata needed to understand the modifications, preventing integrity issues.
Solution Approach 2:
The filter provides feedback through annotations that describe the filtering operations performed on each transaction record. This feedback mechanism enables the ingestion system to compensate for the modifications and maintain accurate data integrity records.
Data Source
AI summary
Here is database replication that, for consistency and security, is based on an innovative unidirectional gateway that modifies and annotates a sequence of human-readable database transaction files. On a communication network, the unidirectional gateway receives a batch of change entries in a database transaction extensible markup language (XML) file. The unidirectional gateway modifies change entry(s). The unidirectional gateway generates and inserts, into the database transaction XML file, metadata that describes the modifications of the change entries. The modified database transaction XML file is converted into a binary file format that for a downstream database and, before or after that conversion, downstream processing of the database transaction file is based on the metadata that describes the modifications of the change entries.


