Integrating LLMs into Cloud Database Platforms via Table-Valued Functions

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current technologies lack the ability to seamlessly integrate large language models (LLMs) into cloud database platforms, leading to increased processing costs and memory usage for natural language processing tasks.

Innovation Solution

The integration of LLMs into cloud database platforms through table-valued functions, allowing users to perform generative natural language processing tasks directly within the platform, thereby eliminating the need for specialized LLMs or application-specific APIs.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If LLM functionality is integrated into cloud database platforms using specialized LLMs or LLM-specific APIs, then natural language processing tasks can be performed, but processing cost and memory usage increase

Engineering Contradiction:

Improvenatural language processing capabilityVSAvoidprocessing cost and memory usage

Core Design Contradiction:

Adaptability or versatilityVSQuantity of substance

Solution Approach 1:

The patent merges LLM functionality directly into the cloud database platform by integrating the LLM engine with the database system. This allows the database platform to perform natural language processing tasks using its existing infrastructure and resources, eliminating the need for separate specialized LLM deployments and reducing overall processing cost and memory usage while maintaining versatility

Inventive Principle:

Principle #5Merging (Combining)

2Adaptability or versatility

If LLM-specific APIs are used to perform natural language processing tasks, then generative text processing can be achieved, but device complexity and ease of operation worsen

Engineering Contradiction:

Improvegenerative text processing capabilityVSAvoidsystem integration complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent creates a universal interface within the cloud database platform that handles both traditional database operations and natural language processing tasks through a unified architecture. The database system can process both structured queries and unstructured text data using the same platform resources, eliminating the need for separate LLM-specific API integrations and reducing system complexity

Inventive Principle:

Principle #6Universality (Multi-functionality)

3Reliability

If specialized LLMs are deployed for natural language processing, then task-specific performance improves, but computing resource consumption increases

Engineering Contradiction:

Improvenatural language processing performanceVSAvoidcomputing resource consumption

Core Design Contradiction:

ReliabilityVSUse of energy by moving object

Solution Approach 1:

The patent enables the cloud database platform to perform natural language processing tasks using its own existing computing infrastructure and resources. The database system leverages its built-in processing capabilities, storage resources, and existing computational frameworks to execute NLP tasks without requiring external specialized LLM deployments, thereby maintaining performance while reducing computing resource consumption

Inventive Principle:

Principle #25Self-service

Data Source

PatentUS20250036670A1Large Language Models in Cloud Database Platforms

Publication Date: 2025.01.30 GOOGLE LLC

AI summary

Aspects of the disclosure are directed to integrating one or more large language models (LLMs) into a cloud database platform, such as a data warehouse. Users of the cloud database platform can provide queries to instruct one or more LLMs to perform generative natural language processing tasks by manipulating or generating text directly in the cloud database platform with a table valued function. Users can provide input to register or generate one or more LLMs of the cloud database platform for performing the natural language processing tasks. Integrating LLMs into the cloud database platform can improve processing capabilities of the LLMs and save computing resources, as specialized LLMs or application-specific API may no longer be necessary.