Remove scale constraints at per-ACM instance level

The scale at the Far Edge is expected to be > 20-50,000 sites and even higher. At such scale, rolling out upgrades is a challenge.

A key requirement to allow for quicker upgrades/updates across a fleet of clusters is support of a high number of concurrent site upgrades/updates.. The current number of 500/ACM instance is considered low by Verizon.

The key requirement is to remove any constraints:

Concurrency to 3000 SNOs/ACM instance

• Phase 1: Upgrade all 3000 SNOs/ACM hub within a maintenance window
  This places a dependency on on upgrade/update times Concurrency/group/batch. The requirement is all 3000 SNOs upgraded within a 4 hrs maintenance window.

• Phase 2: Upgrade all managed clusters in an instance of ACM as a single group
  Match the maximum number of managed clusters supported per ACM instance

• Argocd cluster application shall match the maximum number of managed clusters per ACM instance (currently 3000)
  • See RFE-4130.