Posted On: Jul 30, 2020

Starting today, you can target EC2 Capacity Reservations within an AWS Resource Group using the Resource Group ARN. With this feature, you can now use Capacity Reservations seamlessly for workloads that span Availability Zones and multiple instance types. This feature integrates seamlessly with EC2 Auto Scaling and EC2 Fleet.  

Using AWS Resource Groups, you can organize your Capacity Reservations into logical groups for different workloads, environments, teams, or special events. You can include multiple Capacity Reservations having different attributes (instance type, platform, and Availability Zone) in a single resource group. In addition to launching an EC2 instance by targeting an individual Capacity Reservation, you can now target the instances to a logical group of Capacity Reservations.  

To use this feature, first use the Resource Groups “CreateGroup API” to create a resource group by providing a group name and specifying the configuration. Next, add your Capacity Reservations to that resource group. You can then specify the newly created resource group’s ARN as a target in your launch template. When your EC2 Auto Scaling group or EC2 Fleet launches an instance in a given Availability Zone, the instance will be matched to a Capacity Reservation within the Availability Zone belonging to the group that matches the instance type and platform combination.  

This feature is available in all commercial AWS Regions except Africa(Cape Town), Europe (Milan), Asia Pacific (Osaka-Local) and China Regions. To learn more about this feature, visit this page. To learn more about provisioning and automatically scale instances across purchase options, Availability Zones (AZ), and instance families with EC2 Auto Scaling, visit this blog. To learn more about simplifying the provisioning of Amazon EC2 capacity across different Amazon EC2 instance types, AWS Availability Zones and across On-Demand, RI and Spot purchase models using Amazon EC2 Fleet, visit this blog. To learn more about Amazon EC2 pricing models, visit this page.