hswaw

Solution for scraping workload metrics would be greatly appreciated. This can be achieved either using properly documented per-namespace prometheus instances, or a single per-cluster instance.