The Ultimate Guide to Reserved Instance and AWS Savings Plan

AWS launched Elastic Cloud Compute in August 2006, back then there was only one pricing model. Aiming for flexibility, the pay-per-use model that we come to now call, “On-Demand” was the sole way to purchase the various resources AWS offered. Back then there was only one region and one size but as the instance families and the available regions grew, Amazon rolled out a new pricing model in 2009. Reserved Instances.


RIs saved users billions of dollars with its 1-year or 3-year commitment plans. However, while it did save money, the very concept of RIs was at odds with many of the ideas associated with AWS services. The core promise given to AWS users was flexibility and elasticity of resources.

Limiting yourself to a fixed price resource over time takes away the capacity to utilize any number of services with the possibility of elastic adjustments.


Amazon went through hoops by bringing up new features like selling unwanted RIs and purchasing convertible RIs to fix the lack of freedom its customers have. But such roundabout strategies came with purchase complexities and management complications. Despite all this, RIs were still successful as users were attracted to their discounted pricing. With the cloud market filled with cloud management tools, many enterprises found the management of RIs, Spot and On-Demand services easier to handle. Some organizations paid little heed to the benefits of adjusting specifications strictly for organizational needs.


Introducing AWS Savings Plan


Addressing all of these constraints, AWS announced on the November of 2019, an entirely new discount program that maintains the flexibility and elasticity of its resources but also guarantees lower pricing and longer user commitments, On paper, AWS savings plan is a great introduction, but the two main challenges it faces come from its consumers being unaware of how to take advantage of it or being at odds because they already use Reserved Instances.


Hopefully, by the end of this guide, you’d not only have the full picture on how AWS savings plans work but also what your next would be if you already have other cost-saving plans in hand. 


What is the AWS Savings plan?


With AWS Savings plan, users commit to spending a specific price per hour over a fixed period of time. In return, AWS offers substantial discounts against On Demand rates for EC2 instances and the Fargate container service. Any consumption above the committed amount is charged at On Demand rates.


The purchasing process is far simpler when it comes to AWS Savings plan because of how it allows users to be more flexible with the resource specifications. Purchasing a RI requires about 8 components to be determined whereas you only need 5 for a Savings Plan.

AWS Savings Plan

Reserved Instance

The type of savings plan - EC2 Instance Savings Plan or Compute Savings Plan

The type of reserved Instance- Standard or Convertible

The commitment duration - one year or three years

The commitment duration - one year or three years

Payment - all upfront, partial upfront, or no upfront

Payment - all upfront, partial upfront, or no upfront

The region (Exclusive to EC2 Instance plan)

Region

The instance family( Exclusive to EC2 Instance plan)

Instance family

Size


Operating system


Tenancy



The longer the commitment, the higher the discount. And for customers that choose to pay all upfront will have the benefit of further discounts. Customers can choose the payment commitment such as a minimum of $0.001 per hour per year and then layer multiple Savings Plans together. So for example, a customer can follow up on an instance that is about to have its discount expired with more Savings plans or you can split your savings plans according to which instance requires more utilization.


As mentioned earlier, there are two types of Savings Plan you can choose from. AWS EC2 Instance Savings plan or Compute Savings plan. The two plans, similar to RI and convertible RIs differentiate themselves by offering more discounts for less flexibility. 


AWS EC2 Instance Savings Plan


What is the EC2 Instance Savings Plan?

EC2 Instance Saving plan can offer discounts up to 72% as that of the On-Demand rate depending on the term of commitment, the payment option used, and the instance family has chosen.  The restriction is that you can only use the plan on specific EC2 instances, family and region. You can, however, change the size, OS, and tenancies without losing the plan.



Standard RIs vs EC2 Instance Savings Plan


Both Standard RIs and EC2 Instance Savings Plan are the more restricted options within their respective discount programs. They aim to offer more discounts for fewer liberties.

Categories

EC2 Instance Savings Plan

Standard RIs

Fixed factors

Family and Region-specific

Family and Region-specific

Flexible factors

Can change OS, Size, and Tenancy

Will need to purchase additional RIs for making adjustments

Capacity Reservation

Cannot reserve capacity (You can apply a workaround where you set On-Demand Capacity Reservations and apply a Savings Plan discount on it to gain its benefits.)

Can assign RIs to specific Availability Zones in order to reserve capacity.


AWS Compute Savings Plan


What is the AWS Compute Savings Plan?


For a smaller discount of up to 66%, Compute Savings plan offers more freedom of choice than its counterpart. With the most obvious feature being the plan extension across EC2 and Fargate services. Along with that, you are given freedom of region, instance family and migrated services (for example, if a container service is moved to Fargate).


Convertible RIs vs Compute Savings Plan


Compute Savings plan is a better option than Convertible RIs because of less purchase complexity and improved flexibility which is the main selling point of convertible RIs.

Categories

Compute Savings plan

Convertible RIs

Flexibility

Can change specifications

Can change specifications

Discount allocation

Automatic discounts across all components

Manual adjustments

Capacity reservation

Cannot reserve capacity (You can apply a workaround where you set On-Demand Capacity Reservations and apply a Savings Plan discount on it to gain its benefits.)

Can assign RIs to specific Availability Zones in order to reserve capacity.

Service support

Fargate supported

Elasticache, RDS, and Redshift supported


The drawbacks of Saving Plans

Saving Plans come with its own drawbacks. 

  • You can’t purchase any savings plan for ECS, RDS, Redshift and other services.  
  • There are no options to resell underutilized Savings Plans so your purchases must be careful. 
  • Once you are past your Saving Plan limits, you will be charged On-Demand prices.
  • No capacity reservations
  • The discounts aren’t actually better than Reserved Instances, most of the time, they are the same and sometimes, they are lower.



When to choose between Savings Plan and Reserved instances

The best way to choose between these two options is to answer two questions. How predictable are your resources? And how many varieties of services are you using?


If your resources are not going to be changing its region, size, etc at all then you don’t necessarily need a savings plan. This doesn’t mean you can’t benefit from one but that is under certain circumstances.


If you have an EC2 instance that is predictable, 

  1. You will need reserved instances if its capacities will be reserved 
  2. You will need Savings Plan if it will be running continuously with heavy utilization


Similarly, you can choose a Compute Savings plan instead of Convertible RIs to reduce the management overhead of the resources.


If you are running all your projects on Fargate, then Savings Plan is your definitive choice. However, for customers with Elasticache, RDS, Redshift or container services, you can’t really benefit from the Savings Plan. So all these additional services still require RIs for lower prices.

How to Purchase a Savings Plan- A 5-step Guide

Step 1: Determine Your Resource

Determining your resources is part of finding the specifics of your Savings Plan. For this, you need to be aware of the predictability of your AWS infrastructure. How often will you be changing the scope of your EC2 instances? Will you be utilizing Fargate as a service? What are your budget plans or your workload plans? Also, factor in the waiting period before new purchases. With Savings Plans, there’s no resale option like RIs, so you can add resources to extend the plan but you can’t subtract. 


Step 2: Infrastructure Rightsizing

Rightsizing is important to ensure you don’t spend your savings plans on cloud waste. Stricter management of your resources can avoid unnecessary expenditure. Opt the most effective instance that suits all your technical requirements but nothing more and nothing less. This involves comparing the CPU, memory, disk consumptions and network type all with your requirements. We have a couple of resources that can help you make this decision easier.


Step 3: Analyze your Reservations

This step is crucial if you have existing Reserved Instances. It’s important to optimize your existing RI fleet by modifying Standard Zonal RIs and exchanging Convertible RIs where necessary in order to accurately determine how much Savings Plan coverage is required. 

Optimizing your existing RI fleet will help you better analyze and understand RI usage. 


If your convertible RIs are allocated to dynamic resources that require constant adjustments by different teams, you can exchange them for a Compute Savings Plan to constantly have the benefit of making adjustments with no loss in discounts. Similarly, if your standard zonal RIs are specific to EC2 instances and you have optimized their specs to match your technical requirements, you can slap on an EC2 savings plan on top of them after the reservation term is over.


Step 4: The Purchasing Process

Purchasing can be done in AWS Cost Explorer and the process is similar to purchasing RIs but with the difference being that customers paying partial upfront will have to manually set their payment amount.


Go to Cost Explorer and you will find Savings Plan, click Recommendations for curated suggestions based on your current setup.


Choose the recommendation options as you fit.


The recommendation suggests you will save 40% monthly when committing for a 3-year term. Add the Savings Plan to the Cart.

Click on View Cart.


Review your cart and click submit order. The savings plan will be in effect right away. You can utilize the Cost Explorer’s reports and analysis to review your savings.


Step 5: Monitoring your Savings Plan

Avoid under-using the savings plan you’ve set for yourself by constantly monitoring the resources you have. Savings Plan isn’t a pay-per-use model so despite the level of consumption, you will be charged what you set. And once you’ve used up your consumption rate, your on-demand charges start getting applied. This would facilitate the need to purchase additional plans. Ideally, you should control the consumption rate and seize opportunities to avail of new savings plans.


You can use utilization reports, coverage reports, and the AWS inventory to monitor and track your consumption. Additionally, you can set up budget plans using AWS budgets to ensure your consumption rate doesn’t go past your limit.


Using RIs and Saving Plans together

Purchasing AWS Savings Plan doesn’t exempt you from following cost optimization strategies. Many of the customers that are planning on taking up Savings Plan must have already purchased RIs for their cheaper prices. Trying to abandon your RIs by selling them all and taking up a Savings Plan isn’t an ideal solution for your budget. Contrary to some opinions, Reserved Instances aren’t getting shafted to accommodate Savings Plans. You can use both programs in conjunction to add smarter discounts to your resources.


By using RIs as layers upon your AWS Savings Plan, you have a backup discount program that can reduce cloud costs until the expiration of your reservations. If you have predictable resources that aren’t covered by RIs, applying a Savings Plan on them can be a safe option. 


The Ultimate Guide to Reserved Instance and AWS Savings Plan

Smart Scheduling at your fingertips

Go from simple to smart, real-time AWS resource scheduling to save cost and increase team productivity.

Learn More
More Posts

You Might Also Like

Cloud Automation
New In: No-code cloud management workflows for Azure, VMware & Private Cloud (in addition to AWS)
At TotalCloud, we’ve been enabling workflow-based cloud management for AWS to make it intuitive, accelerated, and no-code. Instead of programming cloud management use cases or depending on siloed solutions, we built out a platform that gives you building blocks to assemble any cloud management solution. 
September 4, 2020
Sayonee
Cloud Computing
List of Essential Kubernetes Tools
Kubernetes is a Container-as-a-Service with tons of unique tools to choose from. External tools play a role in integrating with different systems or maintaining control over the clusters you deploy. Manual health checks and troubleshooting is not ideal to keep a system in full health.This list of tools will provide ample support to your containers and have enough configuration to leave management flexible...
August 12, 2020
Hrishikesh
AWS Use Case Files
TotalCloud Inventory Actions: Giving a new meaning to Cloud Inventory
Learn how the TotalCloud Inventory Dashboard can become equivalent to your cloud provider’s SDK. Carry out any action on any discovered resource with Inventory Actions.
July 30, 2020
Sayonee
AWS Tips & Tricks
AWS Tutorial: Create an AWS Instance Scheduler with Terraform
Terraform is a popular IaaS tool used by many to create, update, and maintain their AWS architecture. If you use Terraform to provision your AWS architecture, you won’t be disappointed with our new AWS tutorial video.We provide you with the means to set up your own instance scheduler from Terraform...
July 20, 2020
Hrishikesh
Cloud Computing
Azure vs AWS: What you need to know
Companies that have jumped the gun with cloud migration during this time of crisis have committed a fatal mistake. The knowledge gap among businesses that seek to migrate is often underestimated, leading to devastating expenditures and operational inefficiencies...
July 15, 2020
Hrishikesh
AWS Use Case Files
Automating Auto Scaling Group Updates
AWS introduced autoscaling to make EC2 cluster scaling easier. We've seen users constantly make changes to their EC2s and put new policies into play. It’s important to update your Autoscaling group with the new instances so that it doesn’t ignore these machines.
July 6, 2020
Hrishikesh