Hardware Redundancy

Abstract

Hardware Redundancy "Those parts of the system that you can hit with a hammer (not advised) are called hardware; those program instructions that you can only curse at are called software." Anonymous Hardware redundancy is achieved by providing two or more physical copies of a hardware component. For example, a system may contain redundant processors, memories, buses, or power supplies. When other techniques, such as use of more reliable components, manufacturing quality control, test, design simplification, etc., have been exhausted, hardware redundancy may be the only way to improve the dependability of a system. For example, in situations in which equipment cannot be maintained, e.g., communication satellites, redundant components allow uninterrupted operating time to be prolonged. Hardware redundancy brings a number of penalties: increase in weight, size, power consumption, cost, as well as time to design, fabricate, and test. A number of choices have to be examined to determine the best way to incorporate redundancy into a system. For example, weight increase can be reduced by applying redundancy to the higher level components. Cost increase can be minimized if the expected improvement in dependability reduces the cost of preventive maintenance for the system. There are three types of hardware redundancy: passive, active, and hybrid. Passive redundancy achieves fault tolerance by masking the faults that occur without requiring any action from the system or an operator. Active redundancy requires a fault to be detected before it is tolerated. After the detection of the fault, the actions of location, containment and recovery are performed to remove the faulty component from the system. Active techniques require that a system is stopped and reconfig-ured to tolerate faults. Hybrid redundancy combines passive and active approaches. Fault masking is used to prevent generation of erroneous results. Fault detection, location, and recovery are used to replace the faulty component with a spare. Hybrid redundancy enables reconfiguration with no system downtime. E. Dubrova, Fault-Tolerant Design,