Serverless

Serverless is a cloud execution model where code runs in stateless functions managed entirely by the cloud provider. Despite the name, servers still exist -- developers just do not provision, configure, or manage them. The provider handles infrastructure, scaling, and availability automatically, charging only for the compute time each function actually uses.

The core unit is a function triggered by an event: an HTTP request, a file upload, a database change, a scheduled timer. Each invocation runs in an isolated environment, completes its work, and shuts down. This means serverless functions scale from zero to thousands of concurrent executions automatically and return to zero when there is no traffic, making idle time genuinely free.

AWS Lambda is the most widely used serverless platform, alongside Google Cloud Functions, Azure Functions, and Cloudflare Workers. Frameworks like Serverless Framework and AWS SAM simplify deployment and local development. Services like Vercel and Netlify have made serverless functions accessible to front-end developers without deep cloud infrastructure knowledge.

The tradeoffs are real and worth understanding before committing. Cold starts -- the latency penalty when a function container has to initialize from scratch -- can add hundreds of milliseconds to response times, though this has improved significantly on major platforms. Debugging distributed function-based architectures is more difficult than debugging a single service. State management becomes an explicit problem since functions are stateless by design. Serverless fits naturally for event-driven workloads, background processing, and APIs with variable traffic, but is awkward for long-running processes or applications with heavy inter-service communication.