Test at extremes (1000x bigger/smaller, instant/year-long) to expose fundamental truths hidden at normal scales. Unsure if approach will scale. Edge cases unclear. Want to validate architecture. "Will this work at production scale?" Need to find fundamental limits.
Test your approach at extreme scales to find what breaks and what surprisingly survives.
Core principle: Extremes expose fundamental truths hidden at normal scales.
| Scale Dimension | Test At Extremes | What It Reveals |
|---|---|---|
| Volume | 1 item vs 1B items | Algorithmic complexity limits |
| Speed | Instant vs 1 year | Async requirements, caching needs |
| Users | 1 user vs 1B users | Concurrency issues, resource limits |
| Duration | Milliseconds vs years | Memory leaks, state growth |
| Failure rate | Never fails vs always fails | Error handling adequacy |
Normal scale: "Handle errors when they occur" works fine At 1B scale: Error volume overwhelms logging, crashes system Reveals: Need to make errors impossible (type systems) or expect them (chaos engineering)
Normal scale: Direct function calls work At global scale: Network latency makes synchronous calls unusable Reveals: Async/messaging becomes survival requirement, not optimization
Normal duration: Works for hours/days At years: Memory grows unbounded, eventual crash Reveals: Need persistence or periodic cleanup, can't rely on memory