On-Demand Instances

aws/billing aws/compute aws/concept

💡 Definition

On-Demand Instances are an EC2 pricing option that lets you pay for compute capacity by the hour or second with no long-term commitments. You have full control over the instance's lifecycle—you decide when to launch, stop, hibernate, or terminate it.

🔑 Key Concepts

No Commitment: You pay only for the time the instances are running.
Flexibility: Ideal for applications with short-term, spiky, or unpredictable workloads that cannot be interrupted.
Highest Cost (per hour): Has the highest per-hour cost compared to Reserved Instances and Spot Instances, but you pay for nothing when it's off.

⚙️ How it Works

When you launch an EC2 instance without specifying a different pricing model, it launches as an On-Demand instance. Billing begins when the instance starts and ends when it is terminated or stopped.

🎯 Use Cases

Development and Testing: For applications being developed or tested on EC2 for the first time.
Unpredictable Workloads: Applications with usage that spikes unpredictably.
Short-Term Projects: For workloads that are not expected to run for a long duration (less than a year).

💰 Pricing Model

Pay-as-you-go: Charged per second (for Linux) or per hour (for Windows) that the instance is in a running state. No upfront payment is required.

📝 Exam Tips (CLF-C02)

Keywords: "No long-term commitment", "Pay by the hour/second", "Unpredictable workloads".
Offers the most flexibility but at the highest hourly price.
Good for applications that need to run without interruption but don't run consistently enough to justify a Reserved Instance or Savings Plan.

See Also: * EC2 * Reserved Instances * Spot Instances * Savings Plans