Facilities should be able to spot network congestion as it arises and keep traffic from piling onto one point once it does. Systems that carry essential communications should keep enough performance headroom that measures used to manage congestion, filtering for instance, leave those services unharmed. The organization should know where its facilities reach their performance ceiling and cap the volume of call or session requests before that ceiling is hit, spreading traffic across distributed facilities where it can. Bearing in mind how large a disruption or disaster might be, it should also weigh distributed processing centres, extra standby facilities and configuration changes that adapt to raise throughput for a time.
The graph holds this control, the 0 it maps to, and the evidence behind each claim, over MCP and REST.