350 rub
Journal Radioengineering №12 for 2019 г.
Article in number:
Managing the reliability of computing systems by optimizing the redundancy architecture
Type of article: scientific article
DOI: 10.18127/j00338486-201912(19)-07
UDC: 62
Authors:

S.A. Skachkov – Dr.Sc.(Eng.), Professor, Main Research Scientist, 

JSC «Concern «Morinsys-Agat» (Moscow)

E-mail: info@concern-agat.ru

A.V. Klyuev – Ph.D.(Eng.), Associate Professor, Senior Research Scientist, 

Vasilevsky Military Academy of Air Defense Forces RF (Smolensk)

E-mail: info@concern-agat.ru

A.A. Kovalev – Senior Research Scientist, 

Vasilevsky Military Academy of Air Defense Forces RF (Smolensk)

E-mail: info@concern-agat.ru

D.A. Moshniakov – Deputy General Director – General Designer, Commercial Director, 

JSC «Concern «Morinsys-Agat» (Moscow)

E-mail: info@concern-agat.ru

Abstract:

For effective confrontation modern SVN SAM (SAM) requires a high level of reliable operation in certain «short» intervals of time when there is a reflection of the air RAID of the enemy. Given the specifics of the combat use of SAMS (AAMS), as well as the intensity of flow of failures and sustained failures, summing the value of the criticality of failures is much greater than the critical steady bounce that is of paramount importance to ensure the reliability of the computing systems in case of hardware failures.

Ensuring the required level of reliability involves the introduction of various kinds of redundancy into the composition of computer systems. The level of reliability of computing systems will depend not only on the amount of resources spent on the introduction of redundancy, but also on how these funds are distributed in the areas of reliability.

To ensure the implementation of the task of improving reliability, a methodological apparatus is proposed that allows to take into account the influence of various types of redundancy on the reliability of computer systems. The main idea of this approach is to represent the decision process in the form of a multi-stage structure, each stage of which is associated with checking the presence of certain properties in a subset of options, which leads to a direct reduction of the original set of options.

To assess the effectiveness of introducing redundancy into the computer system, it is proposed to consider it as a set of separate subsystems, each of which can be implemented in different ways. Failure of any of the subsystems leads to the failure of the entire computer system as a whole. The definition of the system variant is carried out by selecting the method of implementation of each subsystem, delivering the minimum of the target function in the presence of restrictions. At the first stage, possible implementations of the computer system are screened out on the basis of existing limitations, which allows to proceed to the consideration of a smaller number of options. In this case, the introduction of any kind of redundancy, except parametric, is not provided. Next, the possible options for introducing different types of redundancy are considered. The variants of implementation of the computer system selected, taking into account the limitations, are investigated in order to find the extremum of the target function.

Determination of the coefficient of thinning of the flow of failures is proposed to be carried out on the basis of information-time statistical model. This model allows us to solve the problem of quantitative assessment of fault stability in the conditions of different combinations and overlap of redundancy types, to assess the effectiveness of the fault detection system, to determine the coefficient of functional load of the computer system subsystems.

As a result of modeling from the selected variants of implementation of computer systems the variant with a certain redundancy architecture possessing a minimum of target function is chosen. Joint use of the proposed methods will allow to achieve the required level of reliability of computer systems.

Pages: 60-67
References
  1. Zayavka na izobretenie № 2018109822 ot 20.03.2018 g. Sposob obespecheniya nadezhnosti vychislitelnykh sistem pri razlichnykh arkhitekturakh izbytochnosti. (In Russian).
  2. Mikhalevich V.S., Volkovich V.L. Vychislitelnye metody issledovaniya i proektirovaniya slozhnykh sistem. M.: Nauka. Glavred fizmatlit. 1982. (In Russian).
  3. Klyuev A.V. Dis. … kand. tekhn. nauk. Na spetsialnuyu temu. Smolensk: VA VPVO VS RF. 2007. (In Russian).
  4. Skachkov S.A. Metody, modeli i sredstva povysheniya nadezhnosti ZRV voiskovoi PVO v usloviyakh otkazov sboinogo kharaktera vychislitelnykh sistem: Monografiya. Smolensk: VA VPVO VS RF. 2008. 273 s. (In Russian).
Date of receipt: 24 октября 2019 г.