Amazon EC2 Pricing

Different ways of paying for Amazon EC2

Capacity when/how you need it

On-Demand Capacity Reservations

With On-Demand Capacity Reservations, you can reserve compute capacity in a specific Availability Zone. You will have access to Amazon EC2 capacity when and for as long as you need it.

We recommend On-Demand Capacity Reservations for use cases like these:

  • Business-critical events
  • High-availability requirements
  • Disaster recovery
image

Amazon EC2 Capacity Blocks for ML

With Amazon EC2 Capacity Blocks for ML, you can reserve GPU instances in advance for your machine learning (ML) workloads.
We recommend EC2 Capacity Blocks for use cases like these:

  • Train and fine-tune ML models
  • Run experiments and build prototypes
  • Plan for spikes in demand for ML

Learn about EC2 Capacity Blocks pricing

image

Dedicated Hosts

A Dedicated Host is a physical Amazon EC2 server fully dedicated for your use. With Dedicated Hosts, you can use your existing server-bound software licenses. You can purchase them On-Demand (hourly) or as part of Savings Plans.
We recommend Dedicated Hosts for use cases like these:

  • Save on licensing costs
  • Run workloads on dedicated physical servers
  • Offload host maintenance to AWS and control your maintenance event schedules

View Dedicated Hosts pricing

image

Flexible usage with per-second billing

Per-second billing removes the cost of unused compute time from your bill. This particularly helps workloads that run over irregular time periods. Amazon EC2 usage is billed in one-second increments, with a minimum of 60 seconds. The same is true for provisioned storage for Amazon Elastic Block Store (Amazon EBS) volumes. Per-second billing applies to all purchase options. It's available across all Regions and Availability Zones for these instances:

  • Amazon Linux
  • Windows
  • Red Hat Enterprise Linux
  • Ubuntu
  • Ubuntu Pro

For details on related costs, see Amazon EC2 On-Demand Pricing.

Estimate your monthly Amazon EC2 costs