Amazon Elastic Cloud Computing Pricing Guide

What is the Elastic Cloud Computing?

Amazon Elastic Cloud Computing (EC2) is a scalable compute capacity that can adjust to. EC2 instances can be supplied with various CPUs, RAM, storage, and network configuration types for the application’s needs. This service offers nearly limitless capacity for workloads of any size and can enable provisioning in anticipation of unexpected spikes. Working off a pay-as-you-go model, users only have to worry about paying for resources they will be using monthly. Instances can be configured individually for specific workload requirements, and there is a process behind ensuring instances are perfectly adjusted to those requirements.

As an advanced AWS partner, we bring unparalleled expertise to architect, deploy, and optimize cloud solutions tailored to your unique needs. 

EC2 On-Demand

On-Demand lets users choose the instance type and type they want to deploy, with options later to scale up or down the instance for the expected traffic. These instances do not require long-term dedication and can be shut down as quickly as they can be started up. Instances have multiple controls for deployment, termination, hibernation, and rebooting. Pricing is further controlled by the Availability Zones used (AZs), the Regions deployed in, and the operating systems that are used, whether it’s Linux or Windows. Better yet, billing is counted by the hour or by the second. However, this flexibility comes with a larger price tag compared to the other EC2 pricing models.

EC2 Savings Plans

Amazon also offers pricing plans that do come with a dedicated one-to-three year plan. Additionally, these plans do require a consistent use of EC2 resources in dollars per hour. Despite being locked in to AWS’ terms, users can save upwards of 72% of On-Demand prices. During this contract, users can switch EC2 instance types and operating systems of their choice, but are still confined to a specific instance type and AWS Region. However, the degree of these discounts is largely dependent on if the user pays completely upfront, pays at least over half the bill upfront, or waits for automatic billing to start at the end of the month.
In short, EC2 Savings Plans let you:

• Increase or decrease instance size
• Switching from Linux to Windows or vice versa
• Change tenancy between either a dedicated instance or shared instance
• Change Availability Zone

Reserved Instances

Like with the EC2 Savings Plans, Reserved Instances are for one-to-three year commitments that let users save upwards of 72% compared to On-Demand prices. However, Reserved Instances also lets users reserve capacity in an Availability Zone of their choice so they can launch new instances when they need them. Users can also adjust the instance size and the networking of reserved instances. Additionally, there are Convertible Reserved Instances that only allow up to 66% in discounts but allow changes in instance types, tenancy types, and operating systems. These discounts also depend on how much the users are willing to pay upfront.

Spot Instances

This plan is the cheapest, with up to 90% savings compared to On-Demand. The most notable difference is that instances are provided with a surplus of computing resources until AWS needs them for other customers. Prices are not fixed, and instances are instead auctioned off with starting prices constantly being adjusted to current market demand per Availability Zone. Instances will only be usable if the user’s current bid is higher than the current Spot price. If others outbid the user’s maximum price, AWS discontinues the Spot Instance for that user, and the user is swapped over to an EC2 On-Demand instance. These instances are high risk/reward and are not ideal for fault-intolerant applications.

Dedicated Hosts

Dedicated hosts are physical servers dedicated to specifically EC2 the users can rent. This model grants users the resilience and flexibility of AWS with the cost savings of any user-owned server licenses. While the plan bills users using On-Demand hourly rates, purchasing these instances for one or three years will give users up to a 70% discount. Alternatively, users can save up to 72% through the Savings Plans for Dedicated Hosts. This model is designed for companies that want to maintain compliance or minimize hardware sharing as a part of their security measures.

Taking Control of Amazon EC2 Costs

Visibility is the most crucial factor to consider.  AWS has too many moving parts to count, and it is easy to lose track of the associated costs.  AWS’ native tools for evaluating costs, such as AWS Cost Explorer and any usage reports EC2 itself puts out, only look at costs generally.  It certainly doesn’t help that usage reports put all their values in spreadsheets with seemingly endless rows and columns to sift through.  Costs need to be inspected more closely to understand why some parts of the bill bloat out of control.