On-Demand Pricing
Paying for compute capacity by the second or hour with no commitment.
On-Demand pricing lets customers pay for compute capacity — such as EC2 instances or Fargate tasks — by the second or hour with no upfront payment and no long-term contract. Capacity can be launched or terminated at any time, making it the default choice for applications with unpredictable or short-term usage patterns. The key exam nuance is the cost trade-off: On-Demand carries the highest per-unit price among AWS’s compute purchasing options. When a workload runs steadily for a year or more, Reserved Instances or Savings Plans can reduce that cost by up to 72%, making On-Demand the wrong choice for predictable, baseline workloads. Exam questions often test whether candidates can correctly match workload characteristics — spiky, new, or short-lived — to On-Demand, rather than defaulting to it as universally cheapest.
PlayPrepHQ study notes are written and reviewed against primary exam sources. How we create & review content →