Assessment of computing systems reliability by non-parametric method by small samples of operational data
https://doi.org/10.34020/2073-6495-2024-1-010-029
Abstract
A new method for assessing the reliability of small computing systems that allows the generation of only small samples of operational data is proposed in the article. The result of applying the technique is the posterior failure distribution density, on the basis of which various reliability indicators can be calculated. The methodology consists of two stages: the first is the preparation of operational data, including detection of failures using machine learning methods, and the second is the construction of the failure distribution density using the adapted Rosenblatt–Parzen method. Increasing the efficiency of estimates using the proposed method is achieved by taking into account censored data, compensating for the shift of failure distribution densities and finding the optimal smoothing parameter.
About the Authors
V. S. Nikulin
Novosibirsk State University of Economics and Management
Russian Federation
Russian Federation
Nikulin Vladimir S. – Graduate Student
Novosibirsk
A. I. Pestunov
Novosibirsk State University of Economics and Management
Russian Federation
Russian Federation
Pestunov Andrey I. – Candidate of Physical and Mathematical Sciences, Associate Professor, Head of Department
Novosibirsk
References
1. Antonov A.V., Nikulin M.S. Statisticheskie modeli v teorii nadezhnosti: ucheb. posobie [Statistical models in reliability theory: textbook]. Moscow, Abris, 2012. 390 p.
2. Viktorova V.S., Lubkov N.V., Stepanjanc A.S. Analiz nadezhnosti otkazoustojchivyh vychislitel’nyh system [Reliability analysis of fault-tolerant computing systems]. Moscow, IPU RAN, 2016. 117 p.
3. Viktorova V.S., Stepanjanc A.S. Dinamicheskie derev’ja otkazov [Dynamic fault trees], Nadezhnost’ [Reliability], 2011, no. 3, pp. 20–32.
4. Devroj L., D’jorfi L. Neparametricheskoe ocenivanie plotnosti. L1-podhod [Nonparametric density estimation. L1 approach]. Moscow, Mir, 1988. 408 p.
5. Zaharov D.N., Nikulin V.S. Analiz metodov statisticheskoj ocenki jekspluatacionnoj nadezhnosti vychislitel’nyh kompleksov [Analysis of methods for statistical assessment of the operational reliability of computing systems], Naukoemkie tehnologii v kosmicheskih issledovanijah zemli [Science-intensive technologies in space research of the earth], 2020, no. 1, pp. 64–69.
6. Karlov I.A. Metody vosstanovlenija propushhennyh znachenij s ispol’zovaniem instrumentarija Data Mining [Methods for recovering missing values using Data Mining tools], Vestnik Sibirskogo gosudarstvennogo ajerokosmicheskogo universiteta imeni akademika M.F. Reshetneva [Bulletin of the Siberian State Aerospace University named after academician M.F. Reshetneva], 2011, no. 7, pp. 29–33.
7. Matveevskij V.R. Nadezhnost’ tehnicheskih sistem: ucheb. posobie [Reliability of technical systems: textbook]. Moscow, Moskovskij gosudarstvennyj institut jelektroniki i matematiki, 2002. 113 p.
8. Polovko A.M., Gurov S.V. Osnovy teorii nadezhnosti [Fundamentals of reliability theory]. SPb, BHV-Peterburg. 2006. 704 p.
9. Ruzanov P.A. Metodika sozdanija prilozhenij dlja raboty s bazami dannyh s ispol’zovaniem vysokourovnevogo jazyka Python [Methodology for creating applications for working with databases using the high-level Python language]. Cifrovye tehnologii i informacionnaja bezopasnost’ biznes-processov [Digital technologies and information security of business processes]. Sbornik nauchnyh statej po itogam nauchno-prakticheskoj konferencii s mezhdunarodnym uchastiem. Moscow, 2022. Pp. 272–277.
10. Tarasov A.G. Sistema monitoringa vychislitel’nogo klastera rasshirennoj funkcional’- nosti [Computer cluster monitoring system with extended functionality]: diss. … kand. tehn. nauk. Irkutsk, 2011. 145 p.
11. Uspenskij M.B. Razrabotka i issledovanie metodov i modelej obrabotki diagnosticheskoj informacii dlja obnaruzhenija i lokalizacii neispravnostej v sistemah hranenija dannyh [Development and research of methods and models for processing diagnostic information for detecting and localizing faults in data storage systems]. Moscow, 2020. P. 153.
12. Chencov N.N. Ocenka neizvestnoj plotnosti raspredelenija po nabljudenijam [Estimation of the unknown distribution density from observations], Doklady AN SSSR [Reports of the USSR Academy of Sciences], 1962, no. 1 (147), pp. 45–48.
13. Chubukova I.A. Data Mining: ucheb. posobie [Data Mining: textbook]. Moscow, BINOM. Laboratorija znanij, 2008.
14. Shmidt O.O. Obobshhennaja model’ processa vosstanovlenija v teorii nadezhnosti ispol’zovanija informacionnyh tehnologij [Generalized model of the recovery process in the theory of reliability of the use of information technologies]: diss. … kand. fiz.- mat. nauk. Krasnojarsk, 2008. 125 p.
15. Jefron B. Netradicionnye metody mnogomernogo statisticheskogo analiza [Non-traditional methods of multivariate statistical analysis]. Moscow, Finansy i statistika, 1988. 261 p.
16. Foster I. Globus Toolkit Version 4: Software for Service-Oriented Systems. Journal of Computer Science and Technology, 2006, no. 4, pp. 513–520.
17. Hooke R., Jeeves T. “Direct search” solution of numerical and statistical problems. Journal of the ACM, 1961, no. 8, pp. 212–229.
18. Karunamuni R., Alberts T. A Locally Adaptive Transformation Method of Boundary Correction in Kernel Density Estimation. Journal of Statistical Planning and Inference, 2006, vol. 136, no. 9, pp. 2936–2960.
19. Padgett W.J., McNichols D.T. Nonparametric density estimation from censored data. Commun. Statist. Theory and Meth., 1984, vol. 13, no. 13, pp. 1581–1611.
20. Parzen E. On estimation of a probability density function and mode. Annals of Mathematical Statistics, 1962, no. 33, pp. 1065–1076.
21. Rozenblatt M. Remark on some nonparametric estimates of a density function. Annals of Mathematical Statistics, 1956, no. 27, pp. 832–837.
22. Silverman B. Density estimation for statistics and data analysis. London: Chapman & Hall/CRC. P. 45.
23. Wolski R. Developing a Dynamic Performance Information Infrastructure for Grid Systems [Electronic resource]. Available at: https://www.researchgate.net/publication/2527468 (accessed: 04.01.2024).
24. Intellektual’naja sistema tematicheskogo issledovanija naukometricheskih dannyh. Superkomp’juter :«Lomonosov-2» [Intelligent system for thematic research of scientometric data. Supercomputer “Lomonosov-2”]. [Electronic resource]. Available at: https://istina.msu.ru/equipment/card/19828901/ (accessed: 04.01.2024).
Review
For citations:
Nikulin V.S., Pestunov A.I. Assessment of computing systems reliability by non-parametric method by small samples of operational data. Vestnik NSUEM. 2024;(1):10-29. (In Russ.) https://doi.org/10.34020/2073-6495-2024-1-010-029
Email this article 
