* Prometheus proxies through the apiserver to scrape kubelets
* In multi-controller setups, an apiserver must be able to scrape
kubelets (10250) on other controllers
* Change controllers ASG to heterogeneous EC2 instances
* Create DNS records for each controller's private IP for etcd
* Change etcd to run on-host, across controllers (etcd-member.service)
* Reduce time to bootstrap a cluster
* Deprecate self-hosted-etcd on the AWS platform