EFFICIENCY OF REDUNDANT QUERY EXECUTION IN MULTI-CHANNEL SERVICE SYSTEMS
Read the full article
For citation: Bogatyrev V.A., Slastikhin I.A. Efficiency of redundant query execution in multi-channel service system. Scientific and Technical Journal of Information Technologies, Mechanics and Optics, 2016, vol. 16, no. 2, pp. 311–317. doi:10.17586/2226-1494-2016-16-2-311-317
Subject of Research.The paper deals with analysis of the effectiveness of redundant queries based on untrusted computing in computer systems, represented by multi-channel queuing systems with a common queue. The objective of research is the possibility of increasing the efficiency of service requests while performing redundant copies of requests in different devices of a multi-channel system under conditions of calculations unreliability. The redundant service of requests requires the infallibility of its implementation at least in one of the devices.Method. We have considered estimation of the average time spent in the system with and without the use of redundant requests at the presentation of a simple queuing model of the M / M / n type to analyze the effectiveness of redundant service of requests. Presented evaluation of the average waiting time in the redundant queries is the upper one, since it ignores the possibility of reducing the average waiting time as a result of the spread of the probability of time querying at different devices. The integrated efficiency of redundant service of requests is defined based on the multiplicative index that takes into account the infallibility of calculations and the average time allowance with respect to the maximum tolerated delay of service. Evaluation of error-free computing at reserved queries is received at the requirement of faultless execution of at least one copy of the request. Main Results. We have shown that the reservation of requests gives the gain in efficiency of the system at low demand rate (load). We have defined the boundaries of expediency (efficiency) for redundant service of requests. We have shown the possibility of the effectiveness increasing of the adaptive changes in the multiplicity of the reservation of requests, depending on the intensity of the flow of requests. We have found out that the choice of service discipline in information service systems is largely determined by the economic efficiency of information services, estimated by the profit from the error-free provision of information services obtained per time unit (rate of profit). We have demonstrated the existence of an efficiency area of information services at the redundant service of requests under failures and errors. We have determined the flow rate threshold of requests below which a redundant service of requests is appropriate. Practical Relevance The results presented in the paper can be used in the computer systems design including the systems providing information services.
Acknowledgements. The research was carried out as a part of Research Engineering No 610481 "Development of methods and facilities for system engineering design of information and control computer systems with distributed architecture"
1. Aysan H. Fault-Tolerance Strategies and Probabilistic Guarantees for Real-Time Systems. Vasteras, Sweden, Malardalen University, 2012, 190 p.
2. Kopetz H. Real-Time Systems: Design Principles for Distributed Embedded Applications. Springer, 2011, 396 p. doi: 10.1007/978-1-4419-8237-7
3. Sorin D.J. Fault Tolerant Computer Architecture. Morgan & Claypool, 2009, 103 p.
4. Koren I., Krishna C.M. Fault Tolerant Systems. San Francisco, Morgan Kaufmann Publishers, 2009, 378 p.
5. Gatchin Yu.A., Zharinov I.O., Korobeynikov A.G. Mathematical estimation models of information security system infrastructure at the enterprise. Scientific and Technical Journal of Information Technologies, Mechanics and Optics, 2012, no. 2 (78), pp. 92–95. (in Russian)
6. Vorobyov A.I., Kolbanev M.O., Tatarnikova T.M. Assessment of probabilistic-temporal characteristics of the process of I&R service rendering. Journal of Instrument Engineering, 2014, vol. 57, no. 9, pp. 15–18. (In Russian)
7. Shcheglov K.A., Shcheglov A.Yu. The reservation methods capabilities to enhance integral information and operational security level of modern informational systems. Informacionnye Tehnologii, 2015, vol. 21, no. 7, pp. 521–527. (In Russian)
8. Bogatyrev V.A., Bogatyrev A.V. Optimization of redundant routing requests in a clustered real-time systems. Informacionnye Tehnologii, 2015, vol. 21, no. 7, pp. 495–502. (In Russian)
9. Bogatyrev V.A., Bogatyrev A.V. Functional reliability of a real-time redundant computational process in cluster architecture systems. Automatic Control and Computer Sciences, 2015, vol. 49, no. 1, pp. 46–56. doi: 10.3103/S0146411615010022
10. Bogatyrev V.A., Bogatyrev A.V., Bogatyrev S.V. Estimation of reliability of execution of real-time queries. Journal of Instrument Engineering, 2014, vol. 57, no. 4, pp. 46–48.
11. Bogatyrev V.A. Nadezhnost' variantov razmeshcheniya funktsional'nykh resursov v odnorodnykh vychislitel'n. Engineering Simulation, 1997, no. 3, pp. 21–29.
12. Bogatyrev V.A. An interval signal method of dynamic interrupt handling with load balancing. Automatic Control and Computer Sciences, 2000, vol. 34, no. 6, pp. 51–57.
13. Bogatyrev V.A., Bogatyrev A.V., Bogatyrev S.V. Requests redistribution between computing clusters under degradation. Journal of Instrument Engineering, 2014, vol. 57, no. 9, pp. 54–58.
14. Bogatyrev V.A., Bogatyrev S.V., Bogatyrev A.V. Clusters optimization with the limited availability of clusters groups. Scientific and Technical Journal of Information Technologies, Mechanics and Optics, 2011, no. 1 (71), pp. 63–67.
15. Bogatyrev V.A. The reliability evaluation method for computation systems under discontinuity of computer nodes. Instruments and Systems: Monitoring, Control, and Diagnostics, 2006, no. 12, pp. 20–22.
16. Bogatyrev V.A. Reliability of the iultilevel duplicated failure-safe communication system. Herald of computer and information technologies, 2008, no. 4, pp. 24–32.
17. Bogatyrev V.A. Reliability estimation and optimal redundancy in computer cluster systems. Instruments and Systems: Monitoring, Control, and Diagnostics, 2006, no. 10, pp. 18–21.
18. Dudin A.N., Sung B. Unreliable multi-server system with controllable broadcasting service. Automation and Remote Control, 2009, vol. 70, no. 12, pp. 2073–2084. doi: 10.1134/S0005117909120145
19. Lee M.H., Dudin A.N., Klimenok V.I. The SM/V/N queueing system with broadcasting service. Mathematical Problem in Engineering, 2006, vol. 2006, art. 98171. doi: 10.1155/MPE/2006/98171
20. Aliev T.I. Osnovy Modelirovaniya Diskretnykh Sistem [Basics of Discrete Systems Modeling]. St. Petersburg, NRU ITMO Publ., 2009, 363 p.
21. Kleinrock L. Queueing Systems. NY, Wiley Interscience, 1975.
22. Aliev T.I., Rebezova M.I., Russ A.A. Statistical methods for monitoring travel agencies. Automatic Control and Computer Sciences, 2015, vol. 49, no. 6, pp. 321–327. doi: 10.3103/S0146411615060024
23. Cherkesov G.N. Faulttolerance and survivability of critical systems. Design and Technology of Electronic Means, 2015, no. 1, pp. 15–24.
24. Shubinskii I.B. Functional Safety of Information Systems: Analysis Methods. Moscow, Zhurnal Nadezhnost' Publ., 2012, 295 p.