In-Database Distributed Predictive Analytics via Query Decoupling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing information handling systems require specific software development for each data storage platform to perform complex distributed computations, making it difficult to implement predictive analytics across different platforms like Hadoop, Teradata, SQL Server, and Oracle.
Innovation Solution
A distributed analytics system that decouples data storage technology from analytics implementation, allowing predictive modeling algorithms to be integrated within the database for in-database computations, reducing the need for platform-specific code and enabling efficient execution of data-processing intensive algorithms without moving raw data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If specific software code is developed for each data storage platform to perform distributed predictive analytics computations, then the computations can be executed on that specific platform, but the system becomes difficult to apply across different platforms and requires maintaining multiple platform-specific implementations
Solution Approach 1:
The patent introduces an intermediary layer (analytics processor with query interface) that mediates between the user's predictive analytics requests and the various data storage platforms. This intermediary translates high-level analytics queries into platform-specific operations, eliminating the need for platform-specific software development while maintaining compatibility across Hadoop, Teradata, SQL Server, Oracle and other platforms.
Solution Approach 2:
The analytics processor is designed with universal functionality to work with multiple different data storage platforms through a common query interface. Instead of creating separate software for each platform, a single analytics processor can execute predictive analytics computations across diverse platforms, achieving multi-functionality and broad adaptability.
2Productivity
If raw data is moved from storage system to computation system for predictive analytics operations, then the analytics can be performed on the complete dataset, but the data transfer consumes significant time and resources
Solution Approach 1:
Instead of the conventional approach of moving data from storage to computation systems, the patent inverts the workflow by bringing the computation system to the data. The analytics processor executes predictive analytics computations directly on the storage systems, eliminating data transfer and achieving faster analytics execution.
Solution Approach 2:
The patent extracts the computation-intensive predictive analytics operations from the traditional centralized computation system and places them directly at the storage system level. This extraction allows computations to be performed in-place on the stored data, eliminating the need to move large volumes of raw data and significantly reducing transfer time.
Data Source
AI summary
A system, method, and computer-readable medium for performing a distributed analytics operation. The distributed analytics operation uses interface technologies to de-couple an actual data storage technology from an implementation of distributed analytics. Such a distributed analytics operation obviates requirements to deploy specific computer code onto a data storage platform to specifically target that platform for distributed predictive analytics computations.