.png)
How It Works
AWS Graviton processors power a dedicated set of EC2 instance types, identified by a “g” suffix in the instance family name, such as M7g, C7g, and R7g. When a workload runs on a Graviton instance instead of a comparable Intel or AMD instance, it typically consumes fewer compute resources per unit of work. Because Graviton instances carry a lower on-demand price than equivalent x86 types, teams can reduce their EC2 spend by migrating compatible workloads to Graviton without changing the instance size or service tier. The migration requires the application to compile and run on ARM architecture, which most modern runtimes and languages support natively.
Why It Matters for Cloud Cost
EC2 compute is often the largest line item on an AWS bill. Graviton migration is a rate optimization lever, meaning it reduces the per-hour price of a resource rather than reducing the number of resources consumed. For teams already running at efficient utilization levels, Graviton can lower the cost floor without requiring any reduction in capacity. The savings stack on top of commitment-based discounts: a Graviton instance covered by a Compute Savings Plan or EC2 Instance Savings Plan captures both the architectural price reduction and the reservation discount simultaneously.
Usage AI: Usage AI’s Flex Savings Plan covers EC2 instances, including Graviton-based families, delivering 40 to 60% savings versus on-demand rates with $0 upfront and no lock-in risk.