1. Proposed title of this feature request
Separate ETCD instance for Kubernetes events in OpenShift

2. What is the nature and description of the request?
Kubernetes events can be draining on the etcd performance, e.g. when cluster upgrades or node outages are taking place. This is especially true for larger nodes (>250 pods per node) on sub-optimal hardware (no separate NVMe disks for ETCD). As events are not guaranteed and ETCD is often a bottle neck when it comes to the customer infrastructure there should be the possibility to move the events to a separate instance and make room for the core OpenShift functionality.
Public CNCF references are advising a separate ETCD instance as best practice, see KubeCon NA 2023.

3. Why does the customer need this?
Improved scalability and stability.

4. List any affected packages or components.
kube-apiserver (already offers `--etcd-servers-overrides`)
Maybe Openshift apiserver