From the course: Cloud Architecture: Design Decisions

Unlock this course with a free trial

Join today to access over 24,900 courses taught by industry experts.

Advanced architecture scenario: Generative AI computing

Advanced architecture scenario: Generative AI computing

From the course: Cloud Architecture: Design Decisions

Start my 1-month free trial Buy for my team

Advanced architecture scenario: Generative AI computing

“

- In this video we will break down a cloud-based generative AI architecture, specifically, a case study examines enterprise content generation applications using generative AI. For example, a mid-size marketing firm must generate customized content at scale while maintaining brand consistency. They need something that works today, not a science project that takes too long to build. Here's a no-nonsense generative AI cloud architecture. Core components. We have a data layer, AWS S3, which is a data lake, Pinecone, which is a vector database, and Redis for data caching. We have a processing layer, GPT-4 on Azure from OpenAI, which is the primary large language model, SageMaker, which is the custom model, and Kubernetes orchestration. And then we have an integration layer, API Gateway for external access, gRPC for internal communication, and OAuth2 for authentication. The system uses a three-tier architecture with automatic scaling and failover. Data flows from the storage layer through…

Contents