|
所在平台: Udemy |
课程主页: https://www.udemy.com/course/net-aspire-and-genai-develop-distributed-architectures/
课程评论：没有评论
课程名称：.NET Aspire 和 GenAI 开发分布式架构 2025 课程概述：本课程专注于使用 .NET Aspire 框架设计和实施云原生分布式架构，并通过 Microsoft-Extensions-AI 和语义内核（Semantic Kernel）集成生成式人工智能（GenAI）功能。从微服务基础知识到高级 AI 驱动功能，您将获得构建电子商店系统的实践经验，其中商品目录和购物篮微服务协同工作，使用 PostgreSQL、Redis 和 RabbitMQ 进行消息传递。此外，您将学习如何结合智能功能，例如 Q&A 聊天机器人和语义产品搜索，这些功能由 Ollama 的 Llama/Phi 模型和 RAG（检索增强生成）流程提供支持。 在课程中，您将学习到： - **云原生分布式架构基础**：深入了解微服务架构、容器化和十二因素应用方法，学习实践弹性、可扩展性和 DevOps 工作流程的最佳实践。 - **.NET Aspire 框架用于云原生开发**：了解如何利用 .NET Aspire 简化分布式服务的构建，设置新项目、管理配置，并应用日志记录和可观察性等跨领域问题。 - **使用 PostgreSQL 和 RabbitMQ 的商品目录微服务**：在 PostgreSQL 中存储和管理产品数据，向 RabbitMQ 发布集成事件（如产品价格调整）。 - **使用 Redis 的购物篮微服务**：利用 Redis 维护快速的会话数据，当添加商品至购物篮时与商品目录服务同步，并从 RabbitMQ 消费集成事件以保持购物篮价格同步。使用 Keycloak 和 JWT 存取令牌保护购物篮端点。 - **RabbitMQ 的消息和事件驱动模式**：探索发布/订阅模式、交换、路由键以及处理重试的最佳实践，实现稳健的错误处理，确保微服务间的可靠事件驱动通信。 - **部署、安全性和可观察性**：将微服务容器化，通过 Azure Container Apps 部署，使用 azd up 和 azd down 命令进行操作，按照 .NET Aspire 的项目结构进行流线型 CI/CD 工作流程。 - **.NET GenAI 与语义内核简介**：了解生成式人工智能和大语言模型（LLMs）的基础，集成 Microsoft-Extensions-AI 和语义内核以增强 AI 功能。 - **Ollama、Llama 和 Phi 模型的设置**：在本地或通过容器安装和配置 Ollama，直接在 .NET 微服务中运行 Llama 或 Phi 模型进行推理。 - **电子商店中的 GenAI 使用案例**：使用语义内核和提示工程支持的上下文感知 Q&A 聊天机器人，实现 Ollama 的实时响应；利用向量存储（RAG 流程）生成产品数据的嵌入，通过向量数据库检索、排名并提供个性化的产品推荐。 通过本课程，您将构建一个功能齐全的 AI 驱动的电子商店平台，展示与 .NET Aspire 和 GenAI 结合的事件驱动微服务的强大能力。
In this course, we are designing and implementing cloud-native distributed architectures using the.NET Aspire framework, while integrating Generative AI capabilities (GenAI) through Microsoft-Extensions-AI and Semantic Kernel.From microservices fundamentals to Advanced AI-driven features, you'll gain hands-on experience architecting an E-Shop system where Catalog and Basket microservices work in tandem using PostgreSQL, Redis and RabbitMQ for messaging. You'll also discover how to incorporate intelligent features such as Q & A chatbots and semantic product search, powered by Ollama's Llama/Phi models and RAG (Retrieval-Augmented Generation) flows.Throughout the course, you'll learn:Cloud-Native Distributed Architecture EssentialsDive into microservices architecture, containerization, and the Twelve-Factor App methodology.Learn best practices for resiliency, scalability, and DevOps workflows..NET Aspire Framework for Cloud-Native DevelopmentUnderstand how.NET Aspire simplifies building distributed services.Set up new projects, manage configurations, and apply cross-cutting concerns like logging and observability.Catalog Microservice with PostgreSQL and RabbitMQ:Store and manage product data in PostgreSQL.Publish integration events (e.g., ProductPriceChanged) to RabbitMQ.Basket Microservice with Redis:Maintain fast, session-based data using Redis.Syncs with the Catalog service when adding items to the basket.Consume integration events from RabbitMQ to keep basket prices in sync.Secure basket endpoints with Keycloak using JWT Bearer token.Messaging and Event-Driven Patterns with RabbitMQExplore publish/subscribe patterns, exchanges, routing keys, and best practices for handling retries.Implement robust error handling and ensure reliable event-driven communication across microservices.Deployment, Security, and ObservabilityContainerize microservices and deploy them to Azure Container Apps using the azd up and azd down commands.Follow.NET Aspire's project structure for streamlined CI/CD workflows.Introduction to.NET GenAI with Semantic KernelDiscover the foundations of Generative AI and large language models (LLMs).Integrate Microsoft-Extensions-AI and Semantic Kernel to power advanced AI functionalities.Ollama, Llama, and Phi Models SetupInstall and configure Ollama locally or via containers.Run Llama or Phi models for inference directly within your.NET microservices.GenAI Use Cases in E-ShopCustomer Support Q & A Chatbot:Leverage semantic kernel and prompt engineering for context-aware Q & A.Integrate Ollama to deliver real-time responses to users' questions.Product Semantic Search with Vector Store (RAG Flow):Generate embeddings for product data using Ollama's All-MiniLM model.Use a vector database to retrieve, rank, and deliver personalized product recommendations.By the end of this course, you'll have built a fully functional, AI-powered E-Shop platform that demonstrates the power of event-driven microservices coupled with.NET Aspire and GenAI
