Three pumps feed a central manifold. The manifold feeds two facilities, each needing 100% flow. Total demand: 200%. Total max supply: 240%. That extra 40% is your redundancy. It costs money to keep a pump running that you don't "need." Until you need it.
Click a pump to fail it. Drag sliders to adjust capacity. Hit Auto-Fail and watch the system fight to survive. When spare capacity runs out, the next failure takes everything down.
What you just saw
Redundancy is the boring, expensive insurance that nobody values until the first pump dies. With N+1 redundancy (three pumps for two pumps' worth of demand), you survive one failure without anyone noticing. Without it, one failure means everyone scrambles.
A system with no spare capacity is not efficient. It is fragile. The next failure, any failure, causes cascade. That extra pump is not waste. It is the difference between a system that degrades gracefully and one that collapses completely.
Jargon you just learned
- Redundancy
- Extra capacity or components beyond the minimum needed, kept in reserve for when something fails.
- Cascade Failure
- When one failure causes additional failures in connected components, spreading through the system.
- N+1 Redundancy
- Having one more component than the minimum required, so the system survives a single failure.
- Single Point of Failure
- A component whose failure alone causes the entire system to fail. The thing keeping you up at night.
- Load Balancing
- Distributing work across multiple components so no single one is overwhelmed.
- Resilience
- A system's ability to absorb disturbance and continue functioning.
- Graceful Degradation
- When a system loses capability incrementally rather than failing completely. Bend, don't break.