#atom

Computing model that abstracts server management and scales automatically

Core Idea: Serverless computing is a cloud execution model where the cloud provider dynamically manages the allocation and provisioning of servers, allowing developers to build and run applications without concerning themselves with infrastructure management.

Key Elements

Core Characteristics

• No Server Management: Infrastructure abstracted away from developers
• Event-Driven Execution: Functions triggered by events rather than running continuously
• Auto-Scaling: Automatic scaling from zero to peak demand
• Pay-for-Value: Billing based on actual execution time or resources consumed
• Statelessness: Functions don't maintain state between invocations
• Ephemeral Execution: Short-lived function instances

Serverless Categories

• Function as a Service (FaaS): Event-driven code execution
  • AWS Lambda, Azure Functions, Google Cloud Functions
  • IBM Cloud Functions, Cloudflare Workers
• Backend as a Service (BaaS): Third-party application services
  • Authentication (Auth0, Firebase Auth)
  • Databases (Firebase, DynamoDB)
  • Storage (S3, Azure Blob Storage)
  • API Gateways
• Container as a Service (CaaS): Serverless container execution
  • AWS Fargate, Azure Container Instances
  • Google Cloud Run, Knative

Architecture Patterns

• Function Composition: Chaining functions together
• Event-Driven Architecture: Systems responding to events
• Microservices: Decomposing applications into small services
• API-First Design: Building around well-defined APIs
• Backends for Frontends (BFF): Custom backends for specific frontends
• State Machines: Coordinating complex workflows

Development Considerations

• Cold Start Latency: Initial invocation delay
• Execution Timeouts: Function duration limits
• State Management: External state persistence
• Local Development: Tools and emulators
• Observability: Logging, monitoring, tracing
• Testing: Unit, integration, and end-to-end testing approaches

Deployment and Operations

• Infrastructure as Code (IaC): Declarative resource definitions
  • AWS SAM, Serverless Framework, Terraform
• CI/CD Integration: Automated deployment pipelines
• Versioning and Traffic Shifting: Deployment strategies
• Security: IAM roles, permissions, secrets management
• Cost Optimization: Execution time, memory configuration

Common Use Cases

• API Backends: HTTP/REST APIs
• Data Processing: Stream and batch processing
• Automation: Scheduled tasks and webhooks
• IoT Processing: Device data handling
• Web Applications: Dynamic site functionality
• Mobile Backends: Services for mobile apps
• Real-time Processing: Chat, notifications, analytics

Limitations and Challenges

• Vendor Lock-in: Provider-specific implementations
• Complex Debugging: Distributed nature complicates troubleshooting
• Long-running Tasks: Not ideal for extended processing
• Resource Limits: Memory, execution time, concurrency constraints
• Cold Starts: Latency for infrequently used functions

Additional Connections

• Broader Context: Cloud Computing (parent computing model)
• Applications: Event-Driven Architecture (complementary design approach)
• See Also: Cloud-Native Architecture (broader design philosophy)

References

1. "Serverless Architectures on AWS" by Peter Sbarski
2. "Cloud Native Development Patterns and Best Practices" by John Gilbert

#serverless #faas #cloud-computing #architecture

Connections:

Sources:

• From: Hyper-V - Wikipedia