Skip to main content

🧩 Serverless Architecture

✅ Overview​

An architecture that hides server management and is built around event-driven Functions (FaaS).

✅ Problems Addressed​

  • Reduction of infrastructure operation costs.
  • Automation of scaling.
  • Response to irregular load fluctuations.

✅ Basic Philosophy & Rules​

  • Event-driven by FaaS (Lambda / Cloud Functions etc.).
  • Configuration combining managed services.
  • Scale is automatic, pay-as-you-go.
  • Stateless functions are basic.

Conceptual Diagram​

✅ Suitable Applications​

  • Event-driven processing (Log / Notification / Transformation).
  • Tasks with large burst loads.
  • Small-scale API / BFF / Back-office tools.

❌ Unsuitable Cases​

  • Long-running jobs.
  • Stateful processing.
  • Constant large-scale services with high frequency traffic (Cost increase).

✅ History​

  • Spread from AWS Lambda (2014).
  • Developed into architecture combining BaaS / FaaS.

✅ Related Styles​

  • Microservices: Similar philosophy of small granularity and loose coupling.
  • Event-driven: Strongly linked as a starting point of FaaS.

✅ Representative Frameworks​

  • AWS Lambda / API Gateway / DynamoDB / S3
    Most popular Serverless stack.

  • Google Cloud Functions / Cloud Run / Pub/Sub
    Execution environment and event infrastructure requiring no management.

  • Azure Functions
    Good compatibility with .NET ecosystem.

  • Vercel Functions / Netlify Functions
    Serverless configuration close to Web frontend.

✅ Design Patterns Supporting This Style​

  • Observer
    Processing triggered by events (File generation / Message arrival).

  • Command
    Expresses "single operation" per function.

  • Proxy
    Authentication / Authorization / Throttling by API Gateway.

  • Strategy
    Processing branching per event type.

✅ Summary​

Serverless is a powerful structural style of the cloud era realizing
"Scale Automation × Operation Minimization".