AWS offers multiple pricing models for EC2 depending on your workload type, commitment level, and cost tolerance.

Option Cost Savings Commitment Termination Risk Best For
On-Demand None (baseline) None None Unpredictable workloads
Reserved Instances 30-70% 1 or 3 years None Steady, predictable workloads
Savings Plans 30-70% 1 or 3 years None Flexible long-term workloads
Spot Instances Up to 90% None High (can be reclaimed anytime) Fault-tolerant, batch jobs
Dedicated Hosts None Variable None BYOL licenses, compliance
Dedicated Instances None None None Hardware isolation
Capacity Reservations None None None Guaranteed capacity in a specific AZ

1. On-Demand Instances

Pay for compute capacity by the hour or second with no long-term commitment.

Characteristics:

When to Use:

Example:

Running a test server for 2 days
Cost: Pay for exactly 48 hours of usage
No commitment, stop anytime

2. Reserved Instances (RI)

Reserve capacity for 1 or 3 years and get up to 72% discount compared to On-Demand pricing.

Key characteristics: