Server request rate limiting

[tscurtu]

If the Server is overloaded with expensive requests, it should impose backoff responses instead of risk the Kubernetes Server pod being killed due to exceeding its resource limits consumption.