Redundancy(redirected from Redundancy (computing))
Also found in: Dictionary, Thesaurus, Medical, Legal, Financial.
the presence in a technical device of capacities greater than those required to ensure its normal operation. Redundancy is introduced to increase the reliability of a device under various operating conditions or to prevent interference and malfunctions in the transmitting or receiving equipment from affecting the authenticity of the data being transmitted. A particular type of redundancy is equipment redundancy, or reservation.
Other types of redundancy include data redundancy; time redundancy, which is a time reserve for repeated performance of the operation (for example, repeating calculations two or three times in a computer); and power redundancy, which is a power reserve that may be used under more difficult operating conditions or upon aging of a device (for example, by installing a more powerful engine than is necessary under normal operating conditions).
in engineering, an effective method of increasing the reliability of technical apparatus by incorporating an additional number of components and connections over the minimum necessary to perform specified functions under given operating conditions. The components of the minimized structure of an apparatus, which ensures the operational readiness of the apparatus, are called the major components (MC); redundant components (RC) are those used to ensure the operational readiness of the apparatus if the MC’s fail. Redundancy is classified according to a number of indexes, of which the most fundamental are the redundancy level, the redundancy coefficient, the state of the RC’s up to the time they are put into operation, the provision for joint operation of the MC’s and RC’s with a common load, and the means of combining the MC’s and the RC’s.
The redundancy levels are classified as overall redundancy, in which a reserve is provided in case of failure of the whole object, and partial redundancy, in which reserves are provided for individual parts of the object, such as units, subassemblies, and components; composite redundancy features a combination of overall and partial redundancy methods. The redundancy coefficient refers to the ratio of the number of RC’s to the number of MC’s in the apparatus. Single redundancy is called duplication. Depending on the state of the RC’s up to the time they are put into operation, reserves may be classified as loaded (under the same load as the MC’s), reduced (loaded less than the MC’s), and unloaded (under practically no load). The possibility of joint operation of RC’s and MC’s with a common load is determined by whether the components can be connected simultaneously to the load without causing a failure in the reserve group.
Redundancy also depends on the method of combining MC’s and RC’s in the composition of the reserve group. In the permanent connection method, all components—both MC’s and RC’s—are connected to a common load during the entire time the apparatus is operating. In the semipermanent connection method, only components in good working order remain connected to the common load; those that have failed are disconnected. In partially active redundant systems, only MC’s in good working order are connected to the common load at the start of operations; when an MC fails, an RC is connected but the faulty MC is not disconnected. In active redundant systems only MC’s in good working order are connected to the common load at the start of operation; if, however, one of them fails, an RC is connected and the faulty MC is disconnected. The disconnection of faulty MC’s and the connection of RC’s is accomplished either manually or automatically; in the latter case, appropriate apparatus is required, whose reliability must be considered in designing the whole system.
In practice, the use of redundancy is limited by the permissible values of the weight, volume, cost, or other parameters of the apparatus to be reserved. It is therefore necessary to solve an optimization problem for redundancy that has two aspects: providing the maximum values of the reliability indexes for a given value of a limiting factor and providing specified values of the reliability indexes for a minimum value of a limiting factor.
The types of redundancy discussed above pertain to structural redundancy, which is the most common type. Other types of redundancy also exist, including redundancies with respect to the load and time.
REFERENCEKozlov, B. A., and I. A. Ushakov. Kratkii spravochnik po raschetu na-dezhnosti radioelektronnoi apparatury. Moscow, 1966.
V. N. FOMIN
according to Soviet law, one of the grounds for the administration of an enterprise to annul the contract of employment with industrial or nonindustrial workers. Workers may be dismissed for reasons of redundancy because of a reduction in the amount of work of an enterprise or the introduction of various technical and administrative improvements resulting in the elimination of certain jobs, although the volume of work remains unchanged or even increases. Reorganization of an enterprise or institution by merger or division cannot be considered suitable grounds for annulling a contract of employment unless redundancy is evident.
Workers may be dismissed for reasons of redundancy only if it is impossible to transfer them to a new job with their consent. In deciding which workers may remain on the job, preference is given to workers of higher qualification and labor productivity. In case of equal qualification and labor productivity, preference is given to the following categories of workers: those having a family with two or more dependents, those whose families have no other working members with an independent income, those who have worked steadily at the enterprise for an extended period of time, and those suffering a work-related disability or illness. Other categories receiving preferential treatment are indicated in the Labor Code of the RSFSR (art. 34), the Labor Code of the Ukrainian SSR (art. 42), the Labor Code of the Byelorussian SSR (art. 34), and the Labor Code of the Georgian SSR (art. 36).
Redundancy can also be used to detect and recover from errors, either in hardware or software. A well known example of this is the cyclic redundancy check which adds redundant data to a block in order to detect corruption during storage or transmission. If the cost of errors is high enough, e.g. in a safety-critical system, redundancy may be used in both hardware AND software with three separate computers programmed by three separate teams ("triple redundancy") and some system to check that they all produce the same answer, or some kind of majority voting system.
Technically, redundancy is one minus the ratio of the actual uncertainty to the maximum uncertainty. This is the fraction of the structure of the message which is determined not by the choice of the sender, but rather by the accepted statistical rules governing the choice of the symbols in question.
[Shannon and Weaver, 1948, p. l3]