AWS Cost Optimization: Choosing Between Savings Plans Vs. Reserved Instances

Cloud Enablement

successive-savings-plan-reserved-instances

Every hour your team uses fewer cloud resources than the contracted amount is an hour of lost value. And if you do it carefully and end up under-commitment, you’ll pay for the resources your team needs in the highest AWS price tier. Luckily, AWS recognized this pain point and offered cost-saving plans. The two most popular and widely adopted are reserved instances and savings plans. Both offer substantial savings compared to on-demand compute pricing. Though the reserved instances have been there since 2009, savings plans were introduced in 2019. However, the cloud cost management options provided freedom and speed to development but quickly spawned a host of challenges in managing AWS bills. Let’s check the pros and cons of the cost management options and how they can be used in conjunction with each other.

Reserved Instances

AWS Reserved Instances are a way to obtain significant discounts in comparison to EC2 on-demand prices by committing to a certain amount of usage. Instead of paying for compute that you need on an “on-demand” basis, the Reserved instance lets you configure the term length (usually 1 to 3 years) and commit to a stable, discounted rate. Amazon wins because it secures 24×7 EC2 usage, it can estimate capacity, and customers win because they opt for a lower price. Though customers have saved billions, AWS keeps innovating the offering over time and has rolled out options such as RI marketplace, regional RIs, convertible RIs, and instance size flexibility.

More compute options and flexibility have led to complicated processes of choosing the best RI. Given that reserved instances are always bound to a specific instance type. For example, DevOps teams have different instance types, with options spanning storage size, platform, architecture, regions, and availability zones. Finding the best value for the optimal virtual machine configuration becomes difficult.

  • Greater flexibility provided to customers led to an easier process to lower compute spend at scale across different compute services and hundreds of instance options across AWS regions.
  • It also led to the need for an ability to save money while responding to changes in needs without the management overhead.

Types of Reserved Instances

1. Standard Reserved Instances

  • Standard reserved instances come with lucrative discounts but are much more rigid.
  • It allows you to modify the AZ (availability zone), instance size, and network type.
  • It can be sold in the Reserved Instance Marketplace.

2. Convertible Reserved Instances

  • Convertible reserved instances allow you to convert a researched instance for one with a different configuration.
  • It allows you to change the instance family, OS, tenancy, and payment type if the target reserved instance is of equal or greater value.

Savings Plans

Savings Plan is another flexible pricing model from AWS that allows users to save up to 70% on Amazon EC2 and AWS Fargate usage. This allows customers to have continuous use over 1 or 3 years and receive a lower price for that use in return. Every type of compute usage has an on-demand rate and savings plan price. If customers commit to a $10/hour calculated use, they will get the savings plan’s prices on all uses up to $10/hour. Any use beyond the commitment will be charged at the regular on-demand rates.

Types of Savings Plans

1. EC2 Instance Savings Plan

  • EC2 instance savings plan offers the same savings as standard RI (discount up to 72%).
  • These are purchased for a specific area and family (e.g., t3 or m5) AWS automatically applies the savings plan to the family instances in the region regardless of AZ, size, operating system, and tenancy with the most on-demand savings.
  • A 1-year plan is ideal for companies using the latest instance types. 3-year plans will not allow generation up-gradation.

2. Compute Savings Plan

  • Compute savings plans offer a savings rate similar to Variable RI and can reduce spending upto 66 %.
  • CSP savings plans may apply to EC2, ECS, Lambda and Fargate usage regardless of region, instance family, instance size, operating system, or tenancy.
  • AWS automatically applies the savings plan to instances with the most on-demand savings. This requires much less “management” than a convertible RI. For instance, you can change from C4 to M5 instances, shift a workload from EU (Ireland) to EU (London), or move a workload from EC2 to Fargate or Lambda at any time and automatically continue to pay the Savings Plans price.
  • Compute savings plan is suitable for most of the expenses not covered in RI or EC2 Instance Savings plans and workloads that may migrate to Forget. The risk is higher if you spend hourly.

Reserved Instances vs. Savings Plans

Since savings plans have been introduced, the obvious question is whether RIs are still relevant. There have been substantial differences between the two pricing options for some time, creating pros and cons for AWS customers based on their specific use cases. Let’s discuss some:

  • Currently, savings plans can only be applied to EC2 and Fargate, whereas Reserved Instances have wider applications for EC2, RDS, Redshift, and ElastiCache.
  • Convertible Reserved Instances allow the commitment to be extended during the contract term (for example, add additional reservations to cover more EC2 instances) without the need to extend the duration. This is especially helpful when changes are needed, and there is no point in committing to a new term of one to three years. With savings plans, any additions to the original contract are made with a new contract that starts on day 0.
  • EC2 Instance Savings Plan usages will apply in any family, regardless of OS or tenancy. Standard reserved instances can also apply to use in a given instance type family, but this requires Linux and default tenancy.
  • Standard Reserved Instances can be bought and sold on the AWS Marketplace, allowing for more flexibility.
  • Convertible Reserved Instances are within the scope of a specific instance type, OS tenancy, and region. At the same time, Compute Saving Plans will apply to multiple regions for all your types of use.

Cost Explorer

AWS through “My Billing Dashboard” provides an option of Cost Explorer. There are two separate sections in it for SPs and RIs. This is the best way to get started with the cost savings because each section has a recommendations section where you can select the duration and payment options you require. The cost explorer will recommend what to purchase based on the last 7, 30, or 60 days’ usage. So when you get to a point in the project where things are relatively stable, you can look at these recommendations to see just what savings are available.

Two things to note here :

  • The savings plan recommendations are based on a very cautious algorithm as it does not want to recommend something that will not be used. It’s highly likely that more savings plans can be purchased if revisited after a month.
  • If you have accounts linked that have instances that are on a schedule, they will still show up in the savings plan recommendation. Only purchase the amount that will run 24 hours a day. If you dip below your commitment, you will still be charged full price for the hour.

So to recap, Savings Plans and Reserved Instances are a great way of saving money. Still, the headline percentage savings quoted by AWS are only achievable on certain instance types and over long-term deals paying upfront. They also only apply to certain items within AWS; it’s not a blanket x% off the total bill — Load balancers, EIPs, Storage, Other DB types, etc., are all still charged as before. RIs should be used for RDS and very static workloads on EC2. SPs should be used for EC2 otherwise, as there is greater flexibility.

Conclusion

AWS Savings Plan or AWS Reserved Instance will not automatically optimize your cloud bill. Savings plans can help reduce your AWS bill, but you’re still in charge of optimizing the infrastructure.

That’s why rightsizing is such an important initiative. And if you’re running a large cloud environment, you need a solution that automates your rights tasks.

Tracking what resources are running, families, and teams are owned is time-consuming. And trying to understand all the hundreds of EC2 instances AWS offers (with their pricing plans) is no small feat. And it can take you days or weeks to analyze your inventory and usage to know which cases can be downgraded.

But you can also hire an advanced consulting partner or certified AWS solution architects who leverage the expertise and experience to perfect the examples that focus on your business use cases and customize the bill to your best.

 

Schedule a call

Book a free consultation