In this session of #topeaks, we'll talk about container orchestrators, when to use which service or tool, scaling in-out-up-down, node management when scaling, and which metrics we should use for scaling. In the next session, we'll cover more about this awesome topeak! Important additional information on autoscaling ECS and EKS nodes, starting from ECS: AWS offers now an automated way to scale and ECS cluster nodes in and out. The way it works, is by creating a Capacity Provider based on a new or existing Auto Scaling Group, and then enabling "managed scaling" for the same provider. This is all you need to know if you're looking for a quick "FARGATE" experience, without the overhead costs. However, if you're a geek like us, and interested what's going on under the hood, and the way the autoscaler actually creates metrics and calculates every minutes the need to scale in or out, here's the documentation: https://docs.aws.amazon.com/AmazonECS/latest/developerguide/cluster-auto-scaling.html, specifically look for "How cluster Auto Scaling works". As for EKS, things are rather simpler using the OSS Cluster Autoscaler (https://github.com/kubernetes/autoscaler/tree/master/cluster-autoscaler). The documentation for how to deploy it on EKS, with all the required IAM permissions and different configurations, can be found here: https://docs.aws.amazon.com/eks/latest/userguide/autoscaling.html