Virtual Machines (EC2 / Compute Engine)¶

1. Virtual Machine Overview¶

Virtual machines (VMs) are the most fundamental computing resources in the cloud.

1.1 Service Comparison¶

Item	AWS EC2	GCP Compute Engine
Service Name	Elastic Compute Cloud	Compute Engine
Instance Unit	Instance	Instance
Image	AMI	Image
Instance Type	Instance Types	Machine Types
Launch Script	User Data	Startup Script
Metadata	Instance Metadata	Metadata Server

2. Instance Types¶

2.1 AWS EC2 Instance Types¶

Naming Convention: {family}{generation}{attributes}.{size}

Examples: t3.medium, m5.xlarge, c6i.2xlarge

Family	Purpose	Examples
t	General Purpose (Burstable)	t3.micro, t3.small
m	General Purpose (Balanced)	m5.large, m6i.xlarge
c	Compute Optimized	c5.xlarge, c6i.2xlarge
r	Memory Optimized	r5.large, r6i.xlarge
i	Storage Optimized	i3.large, i3en.xlarge
g/p	GPU	g4dn.xlarge, p4d.24xlarge

Key Instance Specifications:

Type	vCPU	Memory	Network	Use Case
t3.micro	2	1 GB	Low	Free tier, development
t3.medium	2	4 GB	Low-Mod	Small apps
m5.large	2	8 GB	Up to 10 Gbps	General purpose
c5.xlarge	4	8 GB	Up to 10 Gbps	CPU intensive
r5.large	2	16 GB	Up to 10 Gbps	Memory intensive

2.2 GCP Machine Types¶

Naming Convention: {series}-{type}-{vCPU-count} or custom

Examples: e2-medium, n2-standard-4, c2-standard-8

Series	Purpose	Examples
e2	Cost-effective General Purpose	e2-micro, e2-medium
n2/n2d	General Purpose (Balanced)	n2-standard-2, n2-highmem-4
c2/c2d	Compute Optimized	c2-standard-4
m1/m2	Memory Optimized	m1-megamem-96
a2	GPU (A100)	a2-highgpu-1g

Key Machine Type Specifications:

Type	vCPU	Memory	Network	Use Case
e2-micro	0.25-2	1 GB	1 Gbps	Free tier
e2-medium	1-2	4 GB	2 Gbps	Small apps
n2-standard-2	2	8 GB	10 Gbps	General purpose
c2-standard-4	4	16 GB	10 Gbps	CPU intensive
n2-highmem-2	2	16 GB	10 Gbps	Memory intensive

2.3 Custom Machine Types (GCP)¶

GCP allows you to specify vCPU and memory individually.

# Create custom machine type
gcloud compute instances create my-instance \
    --custom-cpu=6 \
    --custom-memory=24GB \
    --zone=asia-northeast3-a

3. Images (AMI / Image)¶

3.1 AWS AMI¶

AMI (Amazon Machine Image) Components: - Root volume template (OS, applications) - Instance type, security group defaults - Block device mapping

# Search available AMI (Amazon Linux 2023)
aws ec2 describe-images \
    --owners amazon \
    --filters "Name=name,Values=al2023-ami-*-x86_64" \
    --query 'Images | sort_by(@, &CreationDate) | [-1]'

# Major AMI types
# Amazon Linux 2023: al2023-ami-*
# Ubuntu 22.04: ubuntu/images/hvm-ssd/ubuntu-jammy-22.04-*
# Windows Server: Windows_Server-2022-*

3.2 GCP Images¶

# List available images
gcloud compute images list

# Images from specific project
gcloud compute images list \
    --filter="family:ubuntu-2204-lts"

# Major image families
# debian-11, debian-12
# ubuntu-2204-lts, ubuntu-2404-lts
# centos-stream-9, rocky-linux-9
# windows-2022

4. Creating Instances¶

4.1 AWS EC2 Instance Creation¶

Console: 1. EC2 Dashboard → "Launch instance" 2. Enter name 3. Select AMI (e.g., Amazon Linux 2023) 4. Select instance type (e.g., t3.micro) 5. Create/select key pair 6. Network settings (VPC, subnet, security group) 7. Storage configuration 8. "Launch instance"

AWS CLI:

# Create key pair
aws ec2 create-key-pair \
    --key-name my-key \
    --query 'KeyMaterial' \
    --output text > my-key.pem
chmod 400 my-key.pem

# Create instance
aws ec2 run-instances \
    --image-id ami-0c55b159cbfafe1f0 \
    --instance-type t3.micro \
    --key-name my-key \
    --security-group-ids sg-12345678 \
    --subnet-id subnet-12345678 \
    --tag-specifications 'ResourceType=instance,Tags=[{Key=Name,Value=MyServer}]'

4.2 GCP Compute Engine Instance Creation¶

Console: 1. Compute Engine → VM instances → "Create" 2. Enter name 3. Select region/zone 4. Select machine configuration (e.g., e2-medium) 5. Boot disk (select OS image) 6. Firewall settings (allow HTTP/HTTPS) 7. "Create"

gcloud CLI:

# Create instance
gcloud compute instances create my-instance \
    --zone=asia-northeast3-a \
    --machine-type=e2-medium \
    --image-family=ubuntu-2204-lts \
    --image-project=ubuntu-os-cloud \
    --boot-disk-size=20GB \
    --tags=http-server,https-server

# SSH keys are automatically managed (OS Login or project metadata)

5. SSH Connection¶

5.1 AWS EC2 SSH Connection¶

# Check public IP
aws ec2 describe-instances \
    --instance-ids i-1234567890abcdef0 \
    --query 'Reservations[0].Instances[0].PublicIpAddress'

# SSH connection
ssh -i my-key.pem ec2-user@<PUBLIC_IP>

# Amazon Linux: ec2-user
# Ubuntu: ubuntu
# CentOS: centos
# Debian: admin

EC2 Instance Connect (Browser): 1. EC2 Console → Select instance 2. Click "Connect" button 3. "EC2 Instance Connect" tab 4. Click "Connect"

5.2 GCP SSH Connection¶

# SSH with gcloud (automatic key management)
gcloud compute ssh my-instance --zone=asia-northeast3-a

# Check external IP
gcloud compute instances describe my-instance \
    --zone=asia-northeast3-a \
    --format='get(networkInterfaces[0].accessConfigs[0].natIP)'

# Direct SSH (if key manually registered)
ssh -i ~/.ssh/google_compute_engine username@<EXTERNAL_IP>

Browser SSH: 1. Compute Engine → VM instances 2. Click "SSH" button in instance row 3. Browser terminal opens in new window

6. User Data / Startup Script¶

Scripts that run automatically when instances start.

6.1 AWS User Data¶

#!/bin/bash
# User Data example (Amazon Linux 2023)

# Update packages
dnf update -y

# Install Nginx
dnf install -y nginx
systemctl start nginx
systemctl enable nginx

# Custom page
echo "<h1>Hello from $(hostname)</h1>" > /usr/share/nginx/html/index.html

Specify User Data in CLI:

aws ec2 run-instances \
    --image-id ami-12345678 \
    --instance-type t3.micro \
    --user-data file://startup.sh \
    ...

Check User Data logs:

# Inside instance
cat /var/log/cloud-init-output.log

6.2 GCP Startup Script¶

#!/bin/bash
# Startup Script example (Ubuntu)

# Update packages
apt-get update

# Install Nginx
apt-get install -y nginx
systemctl start nginx
systemctl enable nginx

# Custom page
echo "<h1>Hello from $(hostname)</h1>" > /var/www/html/index.html

Specify Startup Script in CLI:

gcloud compute instances create my-instance \
    --zone=asia-northeast3-a \
    --machine-type=e2-medium \
    --metadata-from-file=startup-script=startup.sh \
    ...

# Or inline
gcloud compute instances create my-instance \
    --metadata=startup-script='#!/bin/bash
    apt-get update
    apt-get install -y nginx'

Check Startup Script logs:

# Inside instance
sudo journalctl -u google-startup-scripts.service
# Or
cat /var/log/syslog | grep startup-script

7. Instance Metadata¶

Query instance information from inside the instance.

7.1 AWS Instance Metadata Service (IMDS)¶

# Instance ID
curl http://169.254.169.254/latest/meta-data/instance-id

# Public IP
curl http://169.254.169.254/latest/meta-data/public-ipv4

# Availability Zone
curl http://169.254.169.254/latest/meta-data/placement/availability-zone

# IAM role credentials
curl http://169.254.169.254/latest/meta-data/iam/security-credentials/<role-name>

# IMDSv2 (recommended - token required)
TOKEN=$(curl -X PUT "http://169.254.169.254/latest/api/token" \
    -H "X-aws-ec2-metadata-token-ttl-seconds: 21600")
curl -H "X-aws-ec2-metadata-token: $TOKEN" \
    http://169.254.169.254/latest/meta-data/instance-id

7.2 GCP Metadata Server¶

# Instance name
curl -H "Metadata-Flavor: Google" \
    http://metadata.google.internal/computeMetadata/v1/instance/name

# External IP
curl -H "Metadata-Flavor: Google" \
    http://metadata.google.internal/computeMetadata/v1/instance/network-interfaces/0/access-configs/0/external-ip

# Zone
curl -H "Metadata-Flavor: Google" \
    http://metadata.google.internal/computeMetadata/v1/instance/zone

# Service account token
curl -H "Metadata-Flavor: Google" \
    http://metadata.google.internal/computeMetadata/v1/instance/service-accounts/default/token

# Project ID
curl -H "Metadata-Flavor: Google" \
    http://metadata.google.internal/computeMetadata/v1/project/project-id

8. Instance Management¶

8.1 Instance State Management¶

AWS:

# Stop instance
aws ec2 stop-instances --instance-ids i-1234567890abcdef0

# Start instance
aws ec2 start-instances --instance-ids i-1234567890abcdef0

# Reboot instance
aws ec2 reboot-instances --instance-ids i-1234567890abcdef0

# Terminate instance (delete)
aws ec2 terminate-instances --instance-ids i-1234567890abcdef0

# Check instance status
aws ec2 describe-instance-status --instance-ids i-1234567890abcdef0

GCP:

# Stop instance
gcloud compute instances stop my-instance --zone=asia-northeast3-a

# Start instance
gcloud compute instances start my-instance --zone=asia-northeast3-a

# Restart instance (reset)
gcloud compute instances reset my-instance --zone=asia-northeast3-a

# Delete instance
gcloud compute instances delete my-instance --zone=asia-northeast3-a

# Check instance status
gcloud compute instances describe my-instance --zone=asia-northeast3-a

8.2 Change Instance Type¶