Adaptive dependability of information management systems

The paper examines the reliability of an information management system as its ability to provide the required services that can be justifiably trusted. It is assumed that the system functions without an operator. The aim is to ensure the dependability of a multimodule control system, when the problem-solving results are affected by failures, faults and errors of problem-solution by the system’s computation modules (CMs). Conventional fault tolerance methods do not provide the desired effect, as even under infinite structural redundancy yet real capabilities of on-line detection of CM failures or faults the system’s dependability is significantly lower than expected. The paper proposes and evaluates the methods of adaptive dependability. They are to ensure the observability of control systems under limited capabilities of component CM operability supervision, as well as achieving the required levels of dependability of information management systems in cases of insignificant float time and structural redundancy. These goals are achieved through active (and automatic) reassignment of the available computational resources for on-line information processing. The methods of adaptive dependability enable – with no interruption of computational processes and while solving real-world problems – timely automatic detection and elimination of failures, faults of CMs and errors in the solution of specified problems through on-line localization of faulty modules and subsequent automatic reconfiguration of the system with the elimination of such modules from operation.

1. For citation: Shubinsky IB, Zamyshliaev AM, Papić L. Adaptive dependability of information management systems. Dependability 2018;4: 3-9. DOI: 10.21683/1729-2646-2018-18-4-3-9

2. Shubinsky IB. Nadiozhnye otkazoustoychivye informatsionnye sistemy. Metody sinteza [Dependable failsafe information systems. Synthesis methods]. Moscow: Dependability Journal; 2016 [in Russian].

3. Kirpichnikov AP, Vasiliev SN. Particular characteristics of today’s microelectronics and matters of highly dependable and secure control systems design. Dependability 2017;3:10-16.

4. Avizienis A, Laprie J-C and Randell B. Dependability of computer systems. Fundamental concepts, terminology and examples. Technical report. LAAS – CNRS; October, 2000.

5. Rus I, Komi-Sirvio S, Costa P. Computer program with insurance of high reliability. Technical report. IFIP WG-10.4; March, 2008.

6. Borisov AB. Bolshoy ekonomicheskiy slovar [Large economic dictionary]. Moscow: Knizhny mir; 2003 [in Russian]. Shäbe H, Shubinsky IB. Limit reliability of structural redundancy. Dependability 2016;1:9-13.

7. Shubinsky IB. Methods of software functional dependability assurance. Dependability 2014;4:95-101.

8. Potapov IV. Issues of software systems dependability. Dependability 2015;1:58-61.

9. Shubinsky IB, Schäbe H. A systematic approach to protection against glitches. Dependability 2014;3:103-107.

10. Shubinsky IB, Shäbe H. On the definition of functional reliability. In: Steenbergen et al., editors. Proceedings of the ESREL 2013, Safety, Reliability and Risk Analysis: Beyond the Horizon. London (UK): Taylor & Francis Group; 2014. pp. 3021-3027. ISBN 978-1-13800123-7.

11. Shubinsky IB. Adaptive fault tolerance in real-time information systems. Life Cycle Engineering and Management. In: Proceedings of ICDQM-2016. Prijevor (Serbia); 29-30 June 2016. pp.3-14.

12. Gnedenko BV, Kovalenko IN. Vvedenie v teoriyu massovogo obsluzhivaniya [Introduction into the waiting theory]. Kiev: Nauka; 1963 [in Russian].