What is Rate Optimisation?
Rate optimization is the practice of reducing cloud costs by lowering the unit price of the services you consume, without necessarily changing the amount of usage. Unlike usage optimization, which focuses on reducing how much you consume, rate optimization focuses on paying less for the same usage.
This is often achieved by selecting the most cost-effective pricing models, commitment plans, and resource types. This are usually in the form of commercial offerings.
It is important to remember that rate optimsiation is only part of the work to maximise the value of your cloud. It will reduce what you pay for a resource, but it won’t reduce the amount of a resource you are using, and therefore doesn’t address waste, over provisioning or enviornmental impact. A combination of rate and usage optimisation will maximise cloud efficiency.
Examples of rate optimisations
- Savings Plans - Offers a rate reduction on compute resources in exchange for 1 or 3 years commitment. Ideal for predicatble long running workloads.
- Reserved Instances - Offers a rate reduction on other resources (e.g. RDS, Opensearch, Elasticache) in exchange for 1 or 3 years commitment. Ideal for predicatble long running workloads.
- Spot Instances - Buying unused EC2 (compute) capacity at up to 90% discount. Ideal for stateless or interruptible workloads.
- Graviton Instances - ARM-based AWS Graviton processors offer better price/performance for many workloads, typically 20-40% cheaper for equivalent performance.
- Region-Based Rates - Running workloads in lower-cost regions can yield significant savings.
- Tiered Storage Rates - Move infrequently accessed data to cheaper storage classes.