Multi-Column Custom Index Encoding for Database Storage Efficiency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional database systems face challenges with exponential growth in database tables, leading to high storage overhead, impractical index management, and reduced memory efficiency due to the need for multiple indexes across various datatypes, which are not scalable or dynamic, and lack linguistic compatibility.
Innovation Solution
A novel technique for dynamically creating and maintaining multi-column custom indexes by encoding datatypes into a common datatype, reducing the number of physical columns and using housekeeping columns to support all access operations, while preserving ordering and linguistic compatibility, allowing for efficient query operations across different languages and datatypes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If conventional database systems create native indexes on all possible ordered datatype combinations, then query speed is improved, but storage overhead increases exponentially
Solution Approach 1:
The patent creates a universal index structure that can handle multiple datatype combinations using a single index rather than requiring separate indexes for each datatype combination. The index is designed to be datatype-agnostic, allowing it to serve multiple query patterns across different datatypes (string, number, date, etc.) without needing to be recreated for each specific combination.
Solution Approach 2:
The patent changes the parameter approach by not pre-defining datatype combinations but instead dynamically determining the appropriate datatype at query time. This allows the system to adapt to different query requirements without committing to a fixed set of datatype combinations, thereby avoiding exponential storage growth.
2Measurement precision
If conventional database systems maintain multiple custom indexes for different datatypes, then data retrieval accuracy is improved, but index management complexity increases
Solution Approach 1:
The patent implements a universal index management system that handles all datatype combinations through a single management interface. Instead of requiring separate management procedures for each datatype combination, the system provides unified creation, maintenance, and optimization operations that work across all datatypes, significantly reducing management complexity.
Solution Approach 2:
The patent enables the index system to automatically determine the appropriate datatype and create the necessary index structures without requiring manual intervention for each datatype combination. The system self-adapts to query requirements by dynamically selecting and creating indexes based on the actual datatype needed, reducing the burden on database administrators.
3Ease of manufacture
If conventional database systems use fixed index structures, then implementation simplicity is maintained, but adaptability to different datatypes is reduced
Solution Approach 1:
The patent introduces dynamic characteristics to the index system by allowing the datatype to be determined at query time rather than being fixed at index creation time. The system can dynamically adapt to different datatype requirements while maintaining a consistent underlying index structure, achieving both simplicity and versatility.
Solution Approach 2:
The patent creates a universal index structure that can serve multiple datatype requirements without requiring separate implementations for each datatype. This single unified approach maintains implementation simplicity while providing broad adaptability across string, number, date, and other datatype combinations.
Data Source
AI summary
A method, system, and apparatus provide for multiple custom fields associated with an application running at a computing device, where the multiple custom fields are received in a specified sort order. The method includes dynamically building multi-column indexes of the multiple custom fields corresponding to multiple intrinsic datatypes stored in multiple custom field columns of a shared table, where the multiple intrinsic datatypes are converted into a generic-indexable datatype to preserve the specified sort order. The method further includes building a sorted index in a specified order in a multi-column indexable table, where the multi-column indexable table includes a partial copy of data from multiple tenants that inhibit the shared table.