Page-Level Schema Transformation for Relational Databases
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing database systems require database objects to be taken offline for schema changes, leading to prolonged downtime, especially for large or critical objects, making it commercially infeasible to transition them to newer versions.
Innovation Solution
A method for performing page-level schema changes on relational database objects by creating a transformation table to map changes from the source object to the target object, allowing data to be moved without unloading and reloading, thus minimizing downtime.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If database objects are taken offline for schema changes, then schema transformation can be completed, but downtime increases and business operations are disrupted
Solution Approach 1:
The system performs preliminary actions by creating a copy of the source database object and generating a transformation table that maps source pages to target pages before the actual schema transformation. This preparation allows the transformation to proceed efficiently without requiring the object to be taken offline for extended periods, thus reducing downtime while ensuring complete schema transformation.
Solution Approach 2:
The system creates a copy of the source database object to work with during the schema transformation process. This copy allows transformations to be performed on page-level structures without affecting the original object's availability for business operations, enabling the transformation to complete while minimizing disruption and downtime.
2Adaptability or versatility
If database objects are unloaded and reloaded for schema changes, then schema version can be updated, but the process becomes prohibitively time-consuming for large objects
Solution Approach 1:
The system segments the database object into page-level units and processes transformations at this granular level rather than requiring complete unloading and reloading of the entire object. The transformation table maps individual pages from source to target, allowing large objects to be transformed incrementally and efficiently, significantly improving transformation speed while maintaining schema version updates.
Solution Approach 2:
The system changes parameters at the page level (such as page size, data set size, piece size, row ID size, segment size, and compression attributes) without requiring complete object unloading. This parameter-level transformation approach allows schema version updates to occur efficiently even for large objects, maintaining high productivity while achieving adaptability.
3Reliability
If complete object transformation is performed, then schema changes are fully implemented, but the object must remain offline throughout the process
Solution Approach 1:
The system performs preliminary actions by creating a transformation table that accurately maps source pages to target pages before the actual transformation. This preliminary mapping ensures transformation accuracy while allowing the process to proceed quickly, reducing the duration the object remains offline while maintaining reliable schema transformation.
Solution Approach 2:
The system enables continuous useful action by allowing the database object to remain online and operational during the schema transformation process. The transformation occurs in the background using page-level operations and transformation tables, ensuring both transformation accuracy and minimal interruption to business operations, thus reducing offline duration while maintaining reliability.
Data Source
AI summary
Methods, devices and systems which facilitate the conversion of database objects from one schema version (e.g., an earlier version) to another schema version (e.g., a newer version) without requiring the objects be unloaded and reloaded are described. In general, data object conversion applies to both table space objects and index space objects. The described transformation techniques may be used to convert any object whose schema changes occur at the page-level.