Serverless

Both AWS Batch and Lambda run your code without making you manage servers — but they are built for very different shapes of work. Knowing which is which saves you from forcing a long job into a service that caps out at 15 minutes. The core difference Lambda Batch Built for Short, event-driven tasks Long, heavy batch jobs Execution time ≤ 15 min Hours or days — no limit Trigger Events (S3, API, schedule) Job submission Resources Up to 10 GB RAM Any EC2/Fargate size, including GPUs Runs on Lambda’s managed runtime EC2 or Fargate (you don’t manage them, but they exist) Scaling Instant, per invocation Queues jobs, then provisions compute Use case Glue code, reactions, APIs Data processing, ML training, simulations Mental model Lambda: "Run this small function NOW, fast....

Serverless is a shift in how you think about compute: instead of provisioning and running servers, you just deploy code. AWS Lambda is the service that pioneered it, and it is the third compute option in the AWS model — alongside EC2 and Fargate. What serverless means A new paradigm — developers don’t manage servers, they just deploy code (functions). It started as FaaS (Function as a Service), pioneered by AWS Lambda....