Cloud Computing Glossary — Key Terms Explained

{
  "Version": "2012-10-17",
  "Statement": [
    {
      "Sid": "ReadOnlyAppBucket",
      "Effect": "Allow",
      "Action": [
        "s3:GetObject",
        "s3:ListBucket"
      ],
      "Resource": [
        "arn:aws:s3:::my-app-bucket",
        "arn:aws:s3:::my-app-bucket/*"
      ]
    }
  ]
}

// This is an IAM policy. It follows the least-privilege
// principle — only read access is allowed.

// Terms used:
// - IAM policy    → JSON rules attached to a role/user
// - Action        → the API operations allowed
// - Resource      → which AWS objects the policy applies to
// - Effect        → Allow or Deny
// - ARN           → Amazon Resource Name (unique identifier)

// Common mistake:
// Using "Action": "s3:*" and "Resource": "*"
// — this grants full access to every bucket, everywhere.
// That's the #1 audit finding in cloud security.

# Terraform — a cloud VPC that demonstrates several
# glossary terms in one file.
resource "aws_vpc" "main" {
  cidr_block           = "10.0.0.0/16"
  enable_dns_support   = true
  enable_dns_hostnames = true

  tags = { Name = "main-vpc" }
}

# Two subnets in different AZs — high availability
resource "aws_subnet" "public" {
  count             = 2
  vpc_id            = aws_vpc.main.id
  cidr_block        = cidrsubnet(aws_vpc.main.cidr_block, 8, count.index)
  availability_zone = data.aws_availability_zones.available.names[count.index]

  tags = { Name = "public-${count.index}" }
}

# Load balancer across both subnets
resource "aws_lb" "app" {
  load_balancer_type = "application"
  subnets            = aws_subnet.public[*].id
}

# Auto-scaling group — matches demand automatically
resource "aws_autoscaling_group" "app" {
  vpc_zone_identifier = aws_subnet.public[*].id
  min_size            = 2
  max_size            = 10
  desired_capacity    = 2
}

# Glossary terms used in this file:
# - VPC       → isolated network
# - Subnet    → IP range within VPC
# - AZ        → physical datacenter
# - Load Balancer → distributes traffic
# - Auto-scaling  → dynamic capacity
# - Terraform → Infrastructure as Code (IaC)

Cloud cost glossary — the terms that decide your bill

Pricing models:
  On-Demand         → pay per hour/second, no commitment
  Reserved Instance → 1–3 year commitment, 30–70% discount
  Savings Plan      → flexible RI variant, discounts by $/hour
  Spot Instance     → 70–90% cheaper, can be interrupted
  Free Tier         → limited free usage per month

Data transfer:
  Egress            → data leaving the cloud (charged)
  Ingress           → data entering the cloud (usually free)
  Inter-AZ transfer → cross-AZ traffic is charged
  CDN egress        → cheaper than direct egress

Monitoring/cost tools:
  Cost Explorer     → AWS dashboard for spend
  Budgets           → alerts when approaching thresholds
  Cost Allocation Tags → tag resources to attribute cost

Common cost mistakes:
  1. Forgotten idle resources (unattached EBS, old snapshots)
  2. Cross-AZ data transfer for high-traffic apps
  3. Over-provisioned instances ("just in case")
  4. NAT Gateway charges for outbound traffic
  5. Non-optimized storage class (S3 Standard for cold data)

Rule of thumb:
  Steady state → Reserved Instances
  Spiky traffic → On-Demand + Auto-scaling
  Batch jobs    → Spot
  Archival      → Glacier / Coldline / Archive tier