I'm looking for something which can allow setup/automation of cloud resource scaling schedules or rules. Ideally, it'd be cross-provider and allow something like:<p>- At 6am, if service tier of resource X is less than Y, increase it to Y<p>- From 8am-1pm, unless usage % of resource X is greater than 25%, decrease service tier to Z<p>- Override rule 2 if it's the weekend..<p>I guess, some system for evaluating and executing scaling rules, based on some criteria, scheduling, and priority.<p>I've done some digging and can't find such a thing. It's obviously possible with scripts and jobs, but I'd like the advantages of some kind of web UI. Anyone using (or built) anything like this?
[I'm assuming you're not managing nodes using a container orchestration tool like Kubernetes.]<p>Have you looked at something like Auto Scaling Groups (AWS) / Autoscaling Groups (GCP)?<p>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].<p>You can also create arbitrary metrics and arbitrary policies for those metrics and scale with a combination of policies. From [2]:<p>> 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.<p>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]).<p>[1]: <a href="https://cloud.google.com/compute/docs/load-balancing-and-autoscaling" rel="nofollow">https://cloud.google.com/compute/docs/load-balancing-and-aut...</a><p>[2]: <a href="https://cloud.google.com/compute/docs/autoscaler/multiple-policies" rel="nofollow">https://cloud.google.com/compute/docs/autoscaler/multiple-po...</a><p>[3]: <a href="https://www.terraform.io/docs/providers/aws/r/autoscaling_policy.html" rel="nofollow">https://www.terraform.io/docs/providers/aws/r/autoscaling_po...</a><p>[4]: <a href="https://www.terraform.io/docs/providers/google/r/compute_autoscaler.html" rel="nofollow">https://www.terraform.io/docs/providers/google/r/compute_aut...</a>
Not cross provider obviously, but EC2 autoscaling has supported this for years: <a href="https://docs.aws.amazon.com/autoscaling/ec2/userguide/schedule_time.html" rel="nofollow">https://docs.aws.amazon.com/autoscaling/ec2/userguide/schedu...</a>
>I'm looking for something which can allow setup/automation of cloud resource scaling schedules or rules.<p>Terraform + chron/some timing script<p>>but I'd like the advantages of some kind of web UI<p>Set up a Jenkins instance on $CLOUD_PROVIDER_OF_YOUR_CHOICE (~20 min) to run Terraform on a specified schedule