Flexible XML Collection Storage in Object-Relational Databases
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current database systems face inefficiencies in storing XML collections due to the limitation of either storing all collection elements out-of-line or in-line based on XML schema requirements, leading to increased storage costs and table proliferation.
Innovation Solution
A mechanism that determines the form of VARRAY storage on an element-by-element basis within an object-relational database system, using mapping annotations and database type considerations to decide between in-line and out-of-line storage for each collection element, allowing for flexible representation and improved query performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If all collection elements are stored out-of-line in a separate table, then query performance and indexing capability are improved, but storage cost increases and table proliferation occurs
Solution Approach 1:
The patent applies local quality by allowing different collection elements within the same XML schema to be stored using different forms (inline or out-of-line) based on their individual characteristics. Specifically, the system evaluates each collection element's database type, occurrence constraints, and query patterns to determine the optimal storage form, rather than applying a uniform storage approach to all elements. This resolves the contradiction by optimizing query performance for elements that benefit from out-of-line storage while using inline storage for elements where it would be unnecessary overhead.
2Stability of the object's composition
If VARRAY storage form is determined at XML schema level, then consistency is maintained, but flexibility and adaptability to different query patterns are reduced
Solution Approach 1:
The patent segments the decision-making process for VARRAY storage by dividing it into two levels: schema-level declarations provide the baseline storage form, while element-level annotations allow specific collection elements to override this baseline when needed. This segmentation enables the system to maintain overall consistency through schema-level rules while providing flexibility for query optimization at the individual element level, resolving the contradiction between consistency and adaptability.
Solution Approach 2:
The patent introduces dynamics by allowing the storage form determination to adapt based on query patterns and access frequencies. The system can dynamically evaluate whether a collection element should use inline or out-of-line storage based on actual usage patterns, rather than being statically determined solely by schema definitions. This dynamic adaptation resolves the contradiction by maintaining schema-level consistency while allowing flexibility when query patterns warrant optimization.
3Ease of operation
If more tables are created for out-of-line VARRAY storage, then indexing and independent querying capability improve, but device complexity increases
Solution Approach 1:
The patent applies local quality by creating out-of-line tables only for specific collection elements that require independent querying capability, rather than creating tables for all collection elements. The system evaluates each element's access patterns and query requirements to determine whether out-of-line storage with separate table creation is necessary, thereby reducing overall device complexity while maintaining ease of operation for elements that need it.
Data Source
AI summary
A database server determines, on an element-level of granularity, what form of VARRAY storage to map collections of elements defined by a XML schema. A collection element may be mapped to an in-line VARRAY or an out-of-line VARRAY. The determination may based on a variety of factors, including the database type mapped to the collection element, database limitations that limit the form storage for certain database types, and annotations (“mapping annotations”) embedded within that XML schema that specifying a database type for database representation of a collection element or a form of VARRAY storage.