Announcing Auto Scaling for Spot fleets
Starting today, you can automatically scale a fleet of Amazon EC2 Spot instances in order to manage application availability and costs during changes in demand based on conditions you define.
Starting today, you can automatically scale a fleet of Amazon EC2 Spot instances in order to manage application availability and costs during changes in demand based on conditions you define.
You can define these conditions to achieve your desired performance and operational efficiency using any published Amazon CloudWatch metric. You can publish custom metrics specific to your application, or use CloudWatch metrics from services such as Amazon EC2 Container Service (ECS) and Amazon Simple Queue Service (SQS). For example, you can scale based on the CPU or memory usage published by Amazon ECS for container-based applications, or the number of messages in your SQS queue for processing batch jobs and web service requests. You can also scale based on instance metrics published for Spot fleets.
Auto Scaling for Spot fleets is now available in all regions where Spot instances are supported via the AWS Management Console, Amazon CloudFormation, and AWS CLI and SDK at no additional charge. To learn more, visit the Amazon EC2 User Guide or read the AWS Blog. To get started, log in to the AWS Management Console to add new scaling policies to your Spot fleets.