Serverless doesn’t mean “no servers.” It means you don’t manage servers. You write a function, deploy it, and the cloud provider runs it for you. You pay only when it executes.

// This is a complete serverless function (AWS Lambda)
export async function handler(event) {
  const name = event.queryStringParameters?.name || 'World';
  return {
    statusCode: 200,
    body: JSON.stringify({ message: `Hello, ${name}!` }),
  };
}

Deploy this, and it handles 0 to 10,000 requests per second automatically. No servers to configure, no scaling to think about, no patches to install.

How it works

1. You write a function
2. You deploy it to a cloud provider
3. A request comes in → the provider spins up your function → runs it → returns the response
4. No requests? Nothing runs. You pay nothing.

The provider handles: servers, operating systems, scaling, load balancing, security patches, availability.

Serverless providers

Provider	Service	Free tier
AWS	Lambda	1M requests/month
Google Cloud	Cloud Functions	2M invocations/month
Cloudflare	Workers	100K requests/day
Vercel	Serverless Functions	Generous hobby tier
Netlify	Functions	125K invocations/month

When to use serverless

Good fit:

• APIs and webhooks
• Cron jobs / scheduled tasks
• Image processing, file handling
• Low-traffic or spiky-traffic apps
• MVPs and side projects (free tier is generous)

Not ideal:

• Long-running processes (most have a 10-30 second timeout)
• WebSocket connections (stateless by design)
• Apps that need persistent in-memory state
• High-throughput, consistent-traffic apps (a regular server is cheaper)

Serverless vs. traditional server

	Serverless	Traditional server
Scaling	Automatic	Manual or auto-scaling config
Cost at zero traffic	$0	$5-50/month minimum
Cost at high traffic	Can get expensive	Predictable
Cold starts	Yes (first request is slower)	No
Deployment	Push a function	Deploy to a server
Maintenance	None	OS updates, security patches

The cold start problem

When a serverless function hasn’t been called recently, the provider needs to spin up a new instance. This “cold start” adds 100ms-2s of latency to the first request. Subsequent requests are fast.

Mitigation: keep functions small, use lightweight runtimes (Node.js, Python), or use providers with minimal cold starts (Cloudflare Workers).

See also: What is CI/CD? | AWS CLI cheat sheet