.png)
How It Works
When you purchase an EC2 Instance Savings Plan or certain Reserved Instances, AWS can apply the discount to instances of different sizes within the same family. For example, a commitment toward an m5.large can shift to cover an m5.xlarge or m5.2xlarge if your workload changes. The discount is applied in normalized units, so AWS converts each instance size to a standard unit of compute and applies the commitment against that total. A larger instance simply consumes more normalized units, and a smaller instance consumes fewer. This means a single commitment can cover a mix of sizes simultaneously, as long as the total normalized usage does not exceed the commitment.
Why It Matters for Cloud Cost
Without instance type flexibility, a commitment purchased for one specific instance size becomes a liability the moment your workload migrates to a different size. Teams that resize EC2 instances for performance reasons, or that run varied workloads across a family, would otherwise face underutilized commitments and full on-demand charges on the new size. Flexibility eliminates that penalty. It makes commitment planning less brittle because finance and engineering teams can right-size instances without abandoning their discount coverage. The result is higher commitment utilization and a lower effective hourly rate across the fleet.
Usage AI: Usage AI’s Autopilot mode purchases and adjusts commitments daily without manual approval, maintaining coverage as your workload and fleet composition shift over time.