Versioned Reference Perspectives for Database Schema Evolution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing database systems face complexity issues when schema changes are required, particularly for interrelated data, as they often lose compatibility with applications and require a translational layer for backwards compatibility, which can be very complex.
Innovation Solution
A method that uses a version identifier in requests to retrieve data based on defined reference perspectives of edges in a database, allowing for reduced complexity by specifying which version of the data structure to use, enabling easy handling of evolving data structures without needing extensive transformations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a translational layer is added to the database to allow backwards compatibility when schema changes are required, then compatibility with applications is maintained, but the complexity of the database increases significantly
Solution Approach 1:
The patent segments the database schema into versioned reference perspectives, where each version is independently defined and can be selected via version identifiers. This allows the database to maintain multiple schema versions without requiring a complex translational layer, as each version is explicitly stored and retrieved as-is rather than being dynamically translated.
Solution Approach 2:
The patent performs preliminary action by pre-defining and storing multiple versions of reference perspectives before schema changes occur. Applications can then directly query specific versions using version identifiers without needing runtime translation, eliminating the need for complex translational layers that would otherwise be required to handle schema evolution.
2Adaptability or versatility
If the schema of the database is changed to evolve data structures, then the database can adapt to new requirements, but compatibility with existing applications is lost
Solution Approach 1:
The patent introduces dynamics by making the schema version selectable through version identifiers in queries. The database can dynamically switch between different reference perspective versions based on the version identifier provided, allowing the same database structure to serve both legacy and new application requirements without losing compatibility.
Solution Approach 2:
The patent performs preliminary action by pre-defining multiple versions of reference perspectives before schema evolution occurs. This allows the database to maintain historical schema versions while introducing new ones, enabling applications to continue working with their expected schemas while the database evolves its data structures.
3Adaptability or versatility
If a translational layer is introduced to maintain backwards compatibility, then application compatibility is preserved, but the effort required for database transformations increases
Solution Approach 1:
The patent segments the schema into discrete, versioned reference perspectives that are stored as-is. This eliminates the need for complex translational layers and reduces transformation effort, as the database simply stores and retrieves complete versioned schemas rather than translating between schemas.
Solution Approach 2:
The patent uses copying by storing complete copies of reference perspectives in different versions. Instead of maintaining a complex translational layer that transforms schemas, the database stores complete copies of each version and retrieves them as needed, significantly reducing the effort required for database transformations while maintaining backwards compatibility.
Data Source
AI summary
As typical databases evolve and the schema defining the stored data changes, difficulties arise in interfacing with the database and compatibility to applications may be lost. Provided for are methods for retrieving data stored in a database using versioned reference perspectives of edges, which define relationships between nodes. The methods provide for backwards-compatibility in accessing node-data stored in accordance with a pre-defined schema based upon a request including a version identifier. Also provided for are backend systems, frontend systems, and industrial machines for the manipulation of work products.