AWS EC2: Crash Course

1. What is EC2?

EC2 = Elastic Compute Cloud.

Simply:

EC2 gives you a virtual server in AWS that you can use to run your application.

Think of it like renting a computer in the cloud.

Instead of:

Your MacBook → run Node.js/Django

you can have:

You choose how powerful the server is and pay for what you use.

2. What is an EC2 Instance?

An instance = your virtual machine/server.

For example:

You can SSH into it:

ssh ubuntu@YOUR_IP

Then it feels almost exactly like using a Linux server.

3. AMI

When creating an EC2 instance, AWS asks you to choose an AMI.

AMI = Amazon Machine Image

Think:

AMI = starting template for your server.

Examples:

Ubuntu 24.04
Amazon Linux
Windows Server

If you choose Ubuntu, AWS creates an Ubuntu server for you.

4. Instance Type

This determines how powerful your server is.

Example:

t3.micro

or

t3.medium

An instance type controls things like:

• CPU
• RAM
• network performance
• sometimes specialised hardware

Think of it like choosing a laptop:

t3.micro     → small laptop
t3.medium    → decent laptop
m7i.large    → more powerful machine
g5.xlarge    → GPU machine

5. EBS

Your EC2 instance needs storage.

AWS commonly uses EBS (Elastic Block Store).

Think:

EBS = hard drive/SSD attached to your EC2 server.

For example:

Your application, files and operating system can live on it.

Important distinction:

EC2 = computer EBS = disk

6. Security Groups 🔐

This is very important for interviews.

A Security Group controls incoming and outgoing network traffic for your EC2 instance.

For example:

Security Group
SSH       22    → your IP
HTTP      80    → everyone
HTTPS     443   → everyone
Node      3000  → nobody

So if your Django app runs:

localhost:8000

but port 8000 isn’t allowed by the Security Group, the internet can’t reach it.

Example

You deploy:

Django → port 8000

You need something like:

Inbound Rules
80    HTTP     0.0.0.0/0
443   HTTPS    0.0.0.0/0
22    SSH      YOUR_IP

Usually you don’t expose 8000 publicly.

Instead:

7. Public IP vs Private IP

An EC2 instance can have:

Public IP

Used to communicate with the server from the internet.

Example:

18.170.xx.xx

Private IP

Used inside the AWS network.

Example:

10.0.2.15

8. Elastic IP

Here’s an annoying AWS detail.

A normal public IP can change when you stop/start an EC2 instance.

An Elastic IP is a static public IPv4 address.

So:

Your server’s public address stays consistent.

9. SSH

You normally connect to a Linux EC2 instance using SSH.

AWS gives you a key pair:

my-server.pem

Then:

chmod 400 my-server.pem
ssh -i my-server.pem ubuntu@18.xxx.xxx.xxx

Now you’re inside the server.

10. Key Pair 🔑

The .pem file is your private SSH key.

Conceptually:

EC2 has the corresponding public key.

Never commit your .pem file to GitHub.

11. VPC

This one sounds complicated but isn’t.

VPC = Virtual Private Cloud

It’s basically:

Your private network inside AWS.

Example:

You control things like:

• IP ranges
• subnets
• routing
• internet access

12. Subnet

A subnet is a range of IP addresses that belong to the same network.

For example:

VPC
10.0.0.0/16

You can divide that network into smaller ranges:

Subnet A: 10.0.1.0/24
Subnet B: 10.0.2.0/24

Each subnet contains a set of possible IP addresses.

So if your EC2 gets:

10.0.1.25

it’s inside Subnet A.

If another EC2 gets:

10.0.2.30

it’s inside Subnet B.

Why divide a network?

Because you may want different groups of machines to be separated and controlled differently.

For example:

So the core definition is:

A subnet is a smaller IP-address range carved out of a larger network.

That’s really what “subnet” means. The public/private distinction comes after understanding this.

Public and private subnets

Once you have the ranges, you separate them by what should be reachable:

• Public subnet → resources that need to be reachable from the internet, e.g. a web server.
• Private subnet → resources you don’t want directly exposed, e.g. a database.

And importantly, a subnet is associated with one Availability Zone.

13. Internet Gateway

A VPC needs a way to communicate with the internet.

That’s what an Internet Gateway (IGW) does.

Think:

14. Route Table

A route table tells AWS:

“Where should network traffic go?”

Example:

Destination       Target
10.0.0.0/16       local
0.0.0.0/0         Internet Gateway

0.0.0.0/0 basically means:

Anywhere else / the internet.

15. Nginx + EC2

A very common deployment looks like this:

Nginx acts as a reverse proxy.

The user doesn’t directly access:

example.com:8000

Instead:

16. EC2 + Docker

You can also run Docker on EC2.

For example:

This is very common.

Your deployment might be:

git pull
docker compose down
docker compose up -d --build

17. EC2 vs S3

Very common interview question.

EC2

Compute

Run application
Run server
Run Docker
Run Python
Run Node

S3

Object storage

Images
Videos
PDFs
Backups
Static files

Think:

EC2 = computer S3 = storage bucket

18. EC2 vs Lambda

Another common interview question.

EC2

You manage the server:

You worry about:

• OS updates
• server configuration
• scaling
• processes

Lambda

AWS manages the servers.

You basically provide:

Function

and AWS runs it when needed.

So:

EC2 → more control Lambda → less infrastructure management

19. Scaling EC2

Suppose your website becomes popular.

One EC2:

Users
 ↓
EC2

Eventually it can’t handle the traffic.

You can have:

             Load Balancer
             /     |     \
            ↓      ↓      ↓
          EC2     EC2    EC2

This is where Auto Scaling Groups come in.

AWS can automatically:

Traffic increases
       ↓
Create more EC2 instances

Then:

Traffic decreases
       ↓
Remove instances

20. Load Balancer

AWS ALB (Application Load Balancer) can distribute requests.

For example:

Request 1 → EC2 #1
Request 2 → EC2 #2
Request 3 → EC2 #3

This gives you better scalability and availability.

21. A Typical Real-World Architecture

For a Django/React application, you might have:

This is the kind of architecture you should be able to explain in an interview.

22. The EC2 Things You MUST Know

If you’re preparing for an interview, remember these:

The mental model 🧠

If you remember only one diagram, remember this:

And traffic:

That’s EC2 in a nutshell.

If you’re learning AWS for a Full Stack/AI interview, the next things I’d learn after EC2 are S3 → RDS → VPC → IAM → ALB → Auto Scaling → CloudWatch → ECS, in roughly that order.