Pricing models
Cloud finance isn't one-size-fits-all. It's about understanding consumption patterns, budgets, and goals. Here is a description about the different pricing models in AWS and what to consider when deciding what pricing model to use.
On-Demand Pricing
Analogy: "Think of On-Demand pricing like paying for a taxi ride. You only pay for the distance you travel, with no commitments or advanced bookings. If you need a ride only occasionally, this might be the best choice for you.”
Scenario: "Imagine your company is launching a new product and expects a surge in website traffic for just a couple of days. For this short duration, On-Demand instances, which require no long-term commitment, would be ideal."
Reserved Instances
AWS Reserved Instances (RIs) is a discount pricing model that enables organizations to save up to 75 % on On-Demand Instances when they purchase them in advance, for a fixed term of one or three years.
RIs enable you to "book" a certain amount of computing power and pay in advance. The three payment options are:
- All Up-front Reserved Instances (AURI)
- Partial Up-front Reserved Instances (PURI)
- No Up-front Reserved Instances (NURI)
Analogy: "Reserved Instances (RIs) are akin to a yearly gym membership. You commit to using the gym (or in this case, specific AWS resources) for an extended period and, in return, receive a discounted rate, much like getting a discount for paying for a full year at the gym upfront."
Scenario: "Suppose your company runs a stable, mature application with consistent usage patterns. You know you'll need specific resources round the clock. Purchasing RIs would be like getting that gym membership; you commit because you know you'll be using the resources consistently."
Savings Plans
With AWS Savings Plans, you do not purchase actual instances but commit to one or three years of On-Demand Instances at a discounted price. This discount can amount to 72 % off the regular On-Demand price.
Analogy: "Savings Plans can be similar to a flexible gym pass, where you commit to a certain number of hours at the gym over a week or month. It doesn't matter which classes or equipment you use, as long as you stick to your committed hours."
Scenario (Compute Savings Plan): "Imagine your company has multiple projects, each with varying requirements. One month you might need more compute power, the next, more memory. Compute Savings Plans offer the flexibility to change the mix of resources as long as you maintain your committed usage level."
Scenario (EC2 Instance Savings Plan): "Your company has a primary application that runs on a specific type of server, but you occasionally need to resize based on demand. EC2 Instance Savings Plans let you commit to a specific server family but allow flexibility in size, ensuring you have power when needed without overcommitting."
Reserved Instances vs. Savings Plans
| AWS Reserved Instances (RIs) | AWS Savings Plans (SPs) | |
|---|---|---|
| Description | Save up to 72% on Amazon EC2 usage by committing to use a set level of compute power in $/hour for 1 or 3 years in a specific AWS Region and instance family. | Save up to 72% on Amazon EC2 usage by committing to use a set level of compute power in $/hour for 1 or 3 years. |
| Types | Convertible Reserved Instances Standard Reserved Instances |
Compute Savings Plans EC2 Instance Savings Plans |
| Potential savings | Up to 66% for Convertible RIs Up to 72% for Standard RIs |
Up to 66% for Compute SPs Up to 72% for EC2 Instances SPs |
| Usage applies to | Amazon EC2 instances: Linux (Instance family + region) Windows (Instance type + region) |
Amazon EC2 instances (Compute SPs apply regardless of instance family, operating system, AWS Region, instance size, or tenancy. EC2 Instance SPs apply to a specific AWS Region and instance family), AWS Fargate, AWS Lambda and Amazon SageMaker. |
| Capacity reservation | Yes, by default for usage within a specific Availability Zone. Convertible RIs can be exchanged or modified. Standard RIs can be modified but not exchanged. | Not by default. But you can reserve capacity through On-Demand Capacity Reservations. |
| Ideal use case | Steady and predictable usage with the same instances. | Usage is mostly steady and may require changes to instances from time to time. |
When is it best to use Reserved Instance vs. Savings Plans?
Factors to Consider: Predictability, Budget, Commitment Length, Cash Flow.
Reserved Instances
Reserved instances enable a higher percentage of savings but are limited in flexibility. Purchase RIs when you:
- Are heavily reliant on EC2 Instances.
- Use applications that require constant, "always-on" power for the better part of one to three years at a time.
- Are confident that your region, instance type, and family won’t change during the contract period.
- Plan to use it with automation logic via 3rd party tools
As a recommendation, Playground Tech suggests implementing RIs for stable workloads, e.g. Windows SQL on EC2, or as a top-up to savings plans while utilizing automation logic via 3rd party tools.
For information on how to buy Reserved Instances, follow this guide.
Savings Plans
Savings Plans offer flexibility, making them ideal for organizations and systems prone to usage changes. They allow exchanges between instance families, operating systems, sizes, tenancies, and regions. Savings Plans are ideal when you:
- Expect that your usage requirements will fluctuate significantly during the contract period.
- Need a setup that automatically adapts to changes in your infrastructure without constant, manual monitoring.
- Look for discounts for serverless use cases with AWS Lambda, AWS Fargate, and SageMaker applications.
- Mix between Amazon EC2, Fargate, and Lambda usage fluctuates.
As a recommendation, Playground Tech suggests implementing savings plans that can cover around 75-80 % of the overall compute usage.
For information on how to buy Savings Plans, follow this guide.