I’ve just got off of a 10-hour flight back from Las Vegas, and during that time, as well as eating a pretty terrible spinach breakfast burrito, I gave some thought to all of the announcements made at re:Invent. In particular, I was impressed by the numerous features that will help customers save money on running .NET core containers. Below is a list of my favourites.

Amazon EKS on AWS Fargate

A Kubernetes cluster is, very basically,  made up of two parts. There is a control plane on one side and worker nodes on the other. This new feature allows you to use Fargate instead of EC2 instances for your worker nodes. Fargate is like a pool of servers that AWS manage on your behalf. Meaning that you can now have an Amazon Elastic Kubernetes Service cluster but no longer have to maintain a single Amazon Elastic Compute Cloud (EC2) instance. I think of this concept as a Serverless Kubernetes.  You can find out more about Amazon EKS on AWS Fargate over on the launch blog.

AWS Fargate Spot

On a similar note, but in the Amazon Elastic Container Service (ECS) service this time, a new capability has been added that lets you use spare capacity in the AWS cloud to run your tasks. So for example, if you have a container that picked some work off a queue and then did some processing, and you had written it to be fault-tolerant (meaning that if it were interrupted, it wouldn’t matter) then you could now use Fargate Spot and save up to 70%.

ECS Cluster Auto Scaling

Sticking on the subject of ECS, a new feature has been added called ECS Cluster Auto Scaling; This allows you to add a Capacity Provider to your cluster that is associated with an EC2 Auto Scaling Group (ASG). The new feature understands and monitors your container workload more closely and can scale quickly and with better accuracy. It’s a container aware scaling mechanism, and it should be of particular value to customers that are running multiple workloads in the same cluster or have workloads that scale-out rapidly

AWS Builders Library

On Thursday, Werner launched The Amazon Builders’ Library. The resource is a collection of detailed articles written by the technical minds behind AWS and Amazon.com. They explain the design decisions that AWS engineers make when optimizing for security, durability, high availability, and performance. They also cover the process of releasing new software to the cloud and maintaining health & high availability after that. All .NET developers can learn something from this library; it contains valuable patterns that you will be able to put to work in your organizations.

That’s it from me, what were your re:Invent highlights? Let me know in the comments, or over on twitter (where you can find me @thebeebs)