The same enterprise, the same failure, two architectures. Switch between them to see what changes in six properties that decide whether the business keeps running.
Continuity
Continuity is a recovery activity. The essential outcome depends on one path, and the plan is to restore that path quickly after it fails.
Continuity is engineered before failure. Every essential outcome keeps more than one live path, and the design question is which outcome must never stop, not which system to restore first.
MOM-310 Continuity & Resilience
Fault isolation
Shared, flat dependencies let a fault travel. A misbehaving component in one corner of the estate can load every component it can reach.
Dependencies are mapped and bounded. Segments contain a fault where it starts, and traffic that crosses a boundary is rate limited, so one failure cannot take the rest of the system with it.
Backups exist on paper but share the dependency that failed: the same control plane, the same site, the same vendor change. Primary and backup fail together.
Redundant paths are diverse and independent, and failover is automatic and tested. A backup only counts if it survives the failure the primary does not.
MOM-310 Continuity & Resilience
Adaptive recovery
Recovery waits on people discovering the problem, searching for its source without a current map, and escalating for authority to act.
Whole-system visibility detects the anomaly, a live dependency map locates it, and containment runs within pre-delegated authority with the evidence sent to the people accountable. Humans can override at any point.
A change, including one made by a vendor or contractor, is approved once and then trusted. Its impact is discovered in production.
Every change has a known blast radius before it lands, is staged and observed, is authorized progressively, and has a tested path back. Third-party changes are governed like internal ones.
Legacy stays on because no one can prove it is safe to turn off. Every year it costs more to run, and adds more to the complexity that makes failures spread.
Technology is retired on purpose, as soon as nothing essential depends on it alone. The capital it consumed is redeployed. Fewer, simpler parts also mean fewer paths for a failure to travel.
MOM-207 Retirement & Decommissioning, MOM-001 Capital Discipline, MOM-007 Capital Redeployment
Side by side
Property
Traditional Operating Model (KTLO)
LNGO architecture
Continuity
Restore after failure
More than one live path to every essential outcome