.png)
How It Works
When you purchase a Reserved Instance with instance flexibility enabled, AWS normalizes each instance size to a unit scale based on a reference point. A larger instance counts as more units, and a smaller one counts as fewer. Your reserved capacity discount then applies automatically to whatever combination of running instances in that family adds up to your purchased unit total. If you originally sized your reservation for one large instance but later run two smaller instances of equivalent total compute, the discount still applies in full. This normalization logic runs continuously in the background with no manual adjustment required. Instance flexibility is available for Linux Reserved Instances under the default tenancy within a region, and applies specifically to EC2 Instance Savings Plans and Regional Reserved Instances on AWS. Also learn Compute vs EC2 Instance Savings Plans.
Why It Matters for Cloud Cost
Workloads change. Teams rightsize instances, migrate to newer generations, or shift capacity between services. Without instance flexibility, a Reserved Instance purchased for a specific instance type goes to waste the moment that exact type is no longer running, forcing teams to either hold underutilized commitments or sell them on the marketplace. With instance flexibility, the discount follows the workload as long as it stays within the same instance family and region. This makes reservation coverage more durable over time and reduces the risk of paying for commitments that no longer match real usage. For finance and engineering teams managing commitments across multiple accounts, instance flexibility is a key lever for keeping utilization rates high without requiring constant manual intervention.
Usage AI: Autopilot purchases and adjusts commitments daily, accounting for instance flexibility when sizing reservations across your workloads so coverage stays optimized without any manual input.