EC2 Fundamentals
Introduction
EC2 (Elastic Compute Cloud) is AWS's virtual machine service. It's how you get compute
resources in the cloud — from tiny instances for testing to GPU monsters for training.
While modern MLOps often uses managed services (SageMaker, EKS), understanding EC2 is
foundational. Kubernetes nodes are EC2 instances. SageMaker training runs on EC2 under
the hood. When something goes wrong, you often need to SSH into an instance to debug.
Why This Matters
As a test automation engineer, you've probably used VMs or containers for test environments.
EC2 is similar but with more configuration options and tighter integration with AWS services.
Understanding EC2 matters for MLOps because:
- You'll choose instance types for different workloads (CPU vs GPU, memory-optimized, etc.)
- You'll configure security groups (firewalls) to control access
- You'll attach IAM roles so instances can access AWS services
- You'll understand what's actually running when you deploy to "the cloud"
Even if you end up using Kubernetes or serverless, EC2 knowledge is the foundation.
Tasks
Hint: t = burstable, m = general purpose, c = compute, r = memory, p/g = GPU
Hint: ssh -i your-key.pem ec2-user@<public-ip>