bramforgelabs.com mental models redundancy backup
Multidisciplinary Thinking · No. 40

Engineering — Redundancy & Backup Systems

Good engineers don't design for the expected load — they design for several times the expected load, because the cost of being wrong is catastrophic.

Charlie Munger · Poor Charlie's Almanack

Munger's core idea: the engineering concept of a margin of safety — building a bridge to hold thirty thousand pounds while only ever declaring its capacity at ten thousand — is one of the most important models in the entire latticework, and it applies just as much to a balance sheet as to a bridge.

— Paraphrased from Charlie Munger, Poor Charlie's Almanack

01 Load → Safety Factor → Margin

Expected Load

The maximum stress the system will realistically face in normal operation.

Safety Factor Applied

Engineers multiply the expected load by a factor — often 3x to 10x — to set the design capacity.

Margin of Safety

The gap between what could break the system and what it's actually built to withstand — room for the unexpected.

02 Single Point of Failure vs. Redundant Design

Resilient Systems

Redundant by Design

  • Independent backup systems that don't share the same failure mode as the primary system.
  • Capacity built well beyond the expected case, so ordinary variation never approaches the limit.
  • Regular testing of the backup itself, not just the primary system.
Fragile Systems

Single Point of Failure

  • One supplier, one server, one key employee, or one source of financing that the whole operation depends on.
  • Backup systems that exist on paper but were never tested under real conditions.
  • Efficiency drives that quietly strip out the very slack that was providing protection.

03 Case Studies

Aviation · Commercial Jets

Triple-Redundant Hydraulics

Commercial aircraft such as the Boeing 747 are built with multiple independent hydraulic systems so that no single failure can disable flight controls, and international ETOPS rules govern how far twin-engine aircraft may fly from a diversion airport based on their redundancy and reliability standards.

Space · Apollo 13, 1970

The Lunar Module as Lifeboat

After an oxygen tank explosion crippled the Apollo 13 command module en route to the Moon, NASA and the crew used the lunar module's separate life-support and propulsion systems — a backup never designed for that purpose — to keep the astronauts alive and return them safely to Earth.

Finance · Berkshire Hathaway

The Cash Redundancy Principle

Berkshire Hathaway has long kept large cash reserves well beyond any foreseeable operating need. Buffett and Munger have repeatedly described this deliberate excess capacity, documented across Berkshire's annual letters and shareholder meetings, as protection against unforeseeable shocks rather than inefficiency.

04 Building In Your Own Redundancy

01

Identify every single point of failure in your business, finances, or plan.

02

Ask what the realistic worst case looks like, then design capacity for several times that.

03

Keep true backups independent — a backup that fails alongside the primary system isn't a backup.

04

Hold a cash or capacity reserve large enough to look excessive in ordinary times.

05

Resist the temptation to strip out redundancy for efficiency once things have gone well for a while.

Munger's core idea: a margin of safety only works if it's there before you need it — the whole point of a backup system is that you build it before, not after, the emergency.
Charlie Munger, Poor Charlie's Almanack