[I'm assuming you're not managing nodes using a container orchestration tool like Kubernetes.] Have you looked at something like Auto Scaling Groups (AWS) / Au

[I'm assuming you're not managing nodes using a container orchestration tool like Kubernetes.] Have you looked at something like Auto Scaling Groups (AWS) / Autoscaling Groups (GCP)? You can customize the auto-scaling policy for how the nodes scale up / down. You can define the auto-scaling policies around load balancing at some level (e.g., CPU utilization, HTTP load balancing, network load balancing) which is discussed in more detail in the docs below [1]. You can also create arbitrary metrics and arbitrary policies for those metrics and scale with a combination of policies. From [2]: > The autoscaler handles multiple policies by calculating the recommended number of virtual machines for each policy and then picking the policy that leaves the largest number of virtual machines in the group. Usually auto-scaling config is tied to a particular service provider; while I haven't done cloud-agnostic auto-scaling policies before personally, I would guess it could be achieved with a bit more effort using something like Terraform (examples: AWS [3], GCP [4]). [1]: https://cloud.google.com/compute/docs/load-balancing-and-autoscaling https://cloud.google.com/compute/docs/load-balancing-and-aut... [2]: https://cloud.google.com/compute/docs/autoscaler/multiple-policies https://cloud.google.com/compute/docs/autoscaler/multiple-po... [3]: https://www.terraform.io/docs/providers/aws/r/autoscaling_policy.html https://www.terraform.io/docs/providers/aws/r/autoscaling_po... [4]: https://www.terraform.io/docs/providers/google/r/compute_autoscaler.html https://www.terraform.io/docs/providers/google/r/compute_aut...