redundancy

Redundancy refers to the inclusion of extra components, capabilities, or information beyond what is strictly necessary for basic operation. In engineering and systems design, redundancy is used to improve reliability, fault tolerance, and availability by ensuring that a failure in one element does not cause total system failure. In information theory and linguistics, redundancy describes repetition or extra data that supports error detection, interpretation, or clarity.

Types include structural or physical redundancy, which duplicates critical components such as parallel machines, duplicate power supplies, or redundant sensors; functional redundancy, which provides alternative methods to achieve the same function; and information redundancy, which adds data for error detection and correction through checksums, parity bits, or error-correcting codes. Diversity redundancy emphasizes independent failure modes by using different technologies or suppliers to avoid common-cause failures.

Applications span aerospace, nuclear safety, industrial control, and other critical infrastructure where failures carry high risks. In computing, redundancy is common in RAID storage, redundant networking paths, failover clusters, and backup power. Data centers employ geographic redundancy to protect against site-specific disasters, while telecommunications networks use redundant routes to maintain service during outages.

Tradeoffs include higher cost, weight, and complexity, plus increased maintenance. Redundancy can also introduce new failure modes if not properly designed or tested. Practitioners seek a balance between resilience and resource use, sometimes using hot, warm, or cold standby and, where appropriate, diverse redundancy to reduce common-cause risks.