Serverless & Microservices with C#: Azure Functions, Docker, Kubernetes & .NET Aspire

Serverless and Microservices Fundamentals

• 💡 Serverless architecture allows developers to focus on business logic without managing infrastructure, with providers like Azure handling server allocation.
• 🚀 The journey of serverless began in 2014 with AWS Lambda, followed by Azure Functions in 2016, offering benefits like autoscaling, pay-as-you-go pricing, and faster development.
• 🧩 Microservices architecture breaks down applications into loosely coupled, independent components, each handling a specific business capability, with serverless being an ideal way to build this.

Azure Functions and Hosting

• ⚡ Azure Functions (Function as a Service) can be written in various languages (C#, JavaScript, Java), integrating with databases, events, workflows, and APIs.
• 🏗️ Hosting plans for Azure Functions include consumption (event-driven, autoscaling, cost-effective) and container app environments, with consumption being a true serverless option.
• 🛠️ Creating an Azure Function involves setting up an Azure account, resource group, function app name, runtime stack (e.g., .NET), and a storage account.

Developing and Deploying Azure Functions

• 💻 Visual Studio is recommended for developing Azure Functions due to its rich features like debugging, though VS Code is also an option.
• 🎯 Azure Functions triggers are essential for execution, with common types including HTTP (for APIs/webhooks), Timer (scheduled execution), Blob Storage (file processing), Queue Storage (message processing), Event Grid, Service Bus, and Cosmos DB.
• 🚀 A simple health check function can be created using an HTTP trigger, deployed to Azure, and accessed via a globally unique URL.

Microservices Architecture and Patterns

• 🧱 Monolithic applications have a single codebase and deployment unit, leading to challenges in scaling and technology integration.
• 🔗 Service-Oriented Architecture (SOA) evolved to address monolith issues with independent modules and standard protocols, but faced scaling challenges.
• 🎯 Microservices architecture addresses SOA's scaling issues by enabling independent scaling, zero downtime, independent teams, and maintenance isolation, often using orchestrators for management.
• ⚠️ Microservices constraints include adhering to domain-driven boundaries, ensuring replicable services, independent teams, well-defined interfaces, standard communication protocols, and interface-only dependencies.
• 🔄 Communication patterns include synchronous (request-response) and asynchronous (message-based), with solutions like output and input message queues to handle challenges like message duplication and ordering.
• 🛡️ Reliability patterns like exponential retry, circuit breaker, and bulkhead isolation are crucial for handling temporary or cascading failures in microservices.

Onion Architecture and Docker

• 🧅 Onion architecture organizes code in layers (Infrastructure, Application, Domain) with dependencies flowing inward, keeping the core business logic independent.
• 📦 Docker containers provide lightweight, efficient virtualization for microservices, isolating applications and their dependencies, offering a solution to the challenges of deploying diverse technology stacks.
• 🐳 Docker images are read-only templates used to create containers, managed via registries like Docker Hub or Azure Container Registry.
• 🏗️ Multi-stage Docker builds create smaller, more secure, and efficient production images by separating build tools from the runtime environment.

Azure Functions Triggers and IoT

• 🌐 HTTP triggers are common for APIs and webhooks, supporting various HTTP methods and authorization levels (anonymous, function, admin).
• 🗄️ Azure SQL triggers invoke functions based on changes (insert, update, delete) in SQL databases, requiring change tracking to be enabled.
• ☁️ Cosmos DB triggers execute functions in response to changes in Cosmos DB collections, ideal for unstructured data but can be expensive.
• 📨 Azure Service Bus triggers (Queues and Topics) enable reliable, decoupled communication between microservices, handling message queuing and publish-subscribe patterns.
• 🌡️ IoT Hub provides a scalable platform for connecting, monitoring, and managing IoT devices, integrating with Azure Functions via Event Hub triggers for real-time data processing.

Microservices in Practice and Kubernetes

• 🗺️ Route planning microservices use RabbitMQ for message brokering and follow Onion architecture, handling user requests and route matching.
• 🌐 Publisher-subscriber patterns with RabbitMQ facilitate communication between microservices, enabling loose coupling and event-driven architectures.
• ☸️ Kubernetes is a distributed orchestration system managing containerized applications across nodes, providing load balancing, autoscaling, service discovery, and health monitoring.
• 📝 YAML is used for defining Kubernetes resources like Pods, Deployments, and Services, requiring specific syntax rules for indentation and structure.
• 📦 Pods are the smallest deployable units in Kubernetes, capable of hosting one or more containers that share storage and network resources.
• 🔄 Deployments manage Pod replicas, enabling scalable and highly available applications with features like rolling updates and rollbacks.
• 🗄️ Persistent Volumes (PVs) and Persistent Volume Claims (PVCs) provide durable storage for containerized applications, ensuring data persistence beyond container lifecycles.