Cloud API Service Integration via Intermediary Architecture
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing methods for providing AI-based API services are costly and time-consuming, especially when built in an on-premise format, and individually managing each service increases operational expenses and development time.
Innovation Solution
A cloud-based API service method and system that integrates multiple AI-based services using a centralized Software as a Service (SaaS) architecture, utilizing common infrastructure and middleware in the control plane area, allowing for efficient management and operation of AI-based API services, including authentication, data analysis, and job processing through GPU-based worker nodes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If AI-based API services are built in an on-premise format, then service reliability can be ensured, but development time and infrastructure cost increase significantly
Solution Approach 1:
The patent introduces a cloud service provider as an intermediary that manages the complex GPU infrastructure and ML model deployment. The cloud provider acts as a mediator between the user and the underlying technical components, allowing users to access AI services without directly managing the infrastructure, thus reducing development time while maintaining service reliability through the provider's established systems.
Solution Approach 2:
The cloud-based API service architecture allows multiple AI services to be provided through a unified platform. The system can serve various AI model types and multiple users simultaneously through a single infrastructure, eliminating the need for separate on-premise setups for each service and reducing overall development time while maintaining reliability through centralized management.
2Adaptability or versatility
If each API service is individually provided, then service flexibility is improved, but operational cost and complexity increase
Solution Approach 1:
The patent merges multiple AI-based API services into a single integrated cloud platform. Instead of managing separate infrastructure for each service, the system combines multiple services under one unified architecture, reducing operational complexity while maintaining the flexibility to provide different AI services through standardized API interfaces.
Solution Approach 2:
The cloud-based platform provides a universal interface that can deliver multiple different AI services to users. The system uses a standardized service architecture where different AI models and functionalities are accessed through common authentication and request handling mechanisms, reducing operational complexity while maintaining service flexibility.
3Ease of manufacture
If cloud-based integrated API service is implemented, then development and operating cost are reduced, but infrastructure sharing complexity increases
Solution Approach 1:
The cloud service provider acts as an intermediary that manages the shared infrastructure complexity. The provider handles the complexities of resource allocation, service isolation, and infrastructure maintenance, allowing developers to focus on building services without worrying about the underlying shared infrastructure, thus reducing development cost while masking infrastructure sharing complexity.
Solution Approach 2:
The system implements self-service mechanisms where the cloud platform automatically manages resource allocation, scaling, and service deployment. The infrastructure sharing is handled automatically by the system without requiring manual intervention to manage the complexity of shared resources, reducing development cost while abstracting away the infrastructure sharing complexity through automated management.
Data Source
AI summary
A method for providing a cloud-based application programming interface (API) service is provided. The method may include issuing an authentication key corresponding to an application for some of a plurality of API services, performing authentication for a first user using the issued authentication key in response to a call for a first API service by the first user, and executing the first API service based on a result of the authentication.