Database Engine Versioning with Memory Buffer Analytics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional database systems face limitations in managing multiple versions of data for analytics, requiring complete data sets and lacking efficient methods for data versioning, analytics integration, and high-speed access to calculated results.
Innovation Solution
The system extends database technology with efficient methods for storing and accessing different data versions, embedding analytics for dynamic recalculations, and providing high-speed access through memory storage, allowing for flexible analytics and data merging across versions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple complete copies of data are stored for each version, then data versioning is supported, but storage space requirements increase significantly
Solution Approach 1:
The patent segments data storage into base data and version-specific differences. Instead of storing complete copies of data for each version, the system stores the base data once and only the differences (changes) for each version. This segmentation allows multiple versions to be maintained with minimal additional storage space, as each version only requires storage for the incremental changes rather than complete data sets.
Solution Approach 2:
The patent implements a nested structure where versioned data is contained within the base data framework. Each version nests its differences within the context of the base data, allowing the system to reconstruct any version by combining the base data with the appropriate version-specific differences. This nesting approach enables efficient storage where versions are embedded within the base structure rather than stored as separate complete copies.
2Adaptability or versatility
If analytics are external to the database engine, then analytics can be performed on database data, but access speed to calculated results is limited by disk storage speed
Solution Approach 1:
The patent introduces an intermediary component - a buffer - positioned between the database engine and analytics operations. This buffer acts as a mediator that holds data in high-speed memory, allowing analytics to access frequently used data without the latency of disk I/O operations. The buffer serves as an intermediate storage layer that decouples the slow disk storage from the fast analytics processing, thereby improving access speed while maintaining the external analytics architecture.
3Speed
If data is stored in server memory for rapid access, then access speed improves, but storage capacity is limited by physical memory and address space
Solution Approach 1:
The patent transitions from a single-dimension storage approach (either disk or memory) to a two-dimensional hierarchical storage architecture. Data is organized across two dimensions: fast memory storage for actively accessed data and slower disk storage for archival and less frequently accessed data. This dimensional change allows the system to leverage both the speed of memory and the capacity of disk storage, effectively breaking the trade-off between access speed and storage capacity by operating in multiple storage dimensions simultaneously.
Data Source
AI summary
A system for calculating analytics uses a relational database to store inputs, calculates results, and stores them in cache. The system also includes an access layer that provides a unified view of the data in server. A dynamic access layer is generated at runtime to run an analytic to provide a flexible framework for creating business logic.