Database Partitioning for Writable Extensions to Read-Only Replicas
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing database replication methods restrict clients from directly augmenting or appending data to replicated database portions, leading to inefficiencies such as the need for additional data structures, complex join operations, and cumbersome life cycle management.
Innovation Solution
Implementing writable partitions within read-only replicated database tables, allowing clients to directly append or integrate additional data without creating new structures, thereby reducing the need for complex join operations and enhancing data manipulation capabilities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If replicated database portions are designated as read-only, then data integrity and consistency are maintained, but clients cannot directly augment or append additional data to the replicas
Solution Approach 1:
The database table is segmented into multiple partitions, where each partition can have different access permissions. Some partitions are read-only (replicated portions) while others are writable (local extensions), allowing simultaneous data integrity protection and local data augmentation
Solution Approach 2:
Different partitions within the same database table have different quality attributes regarding write access. The read-only partitions maintain replicated data integrity, while writable partitions allow local data augmentation, creating local quality variations within the unified table structure
2Adaptability or versatility
If new database structures are created to store additional data, then clients can enrich replicated data, but storage overhead increases due to foreign references
Solution Approach 1:
The patent merges the replicated database table with local extension data by adding writable partitions directly to the existing table structure. This consolidation eliminates the need for separate database structures and foreign reference relationships, reducing storage overhead while maintaining data enrichment capability
3Ease of operation
If join operations are used to query replicated data and new structures, then comprehensive data access is achieved, but computational resources and time requirements grow polynomially
Solution Approach 1:
By merging replicated and local data into a single unified table structure with multiple partitions, the patent eliminates the need for join operations. Queries can directly access both replicated and local data without polynomial growth in computational resources, maintaining comprehensive data access while improving query efficiency
4Ease of manufacture
If separate structures are used for new data, then data organization is achieved, but life cycle management becomes complex for identifying and deleting records
Solution Approach 1:
The patent combines replicated and local data into a single table structure with partitioned organization. This unified structure simplifies life cycle management because records can be identified and deleted using a single primary key system without complex foreign reference tracking across multiple separate structures
Data Source
AI summary
Techniques for using writeable partitions with read-only replicated records include one or more non-transitory computer-readable media storing program instructions that, when executed by one or more processors of an authorization server, cause the one or more processors to perform a method including receiving a replica of a portion of a database, storing the replica of the portion of the database in a read-only partition, adding a field to the portion of the database in a writeable partition, and writing data to the added field.