Скулиш, М. А.Суліма, С. В.2015-12-232015-12-232015Скулиш М. А. Організація управління в багатоетапних системах масового обслуговування / М. А. Скулиш, С. В. Суліма // Вісник Національного університету "Львівська політехніка". Серія: Радіоелектроніка та телекомунікації : збірник наукових праць. – 2015. – № 818. – С. 80–85. – Бібліографія: 7 назв.https://ena.lpnu.ua/handle/ntb/30931Розглянуто проблеми систем з кількома етапами, кількома типами ресурсів, кількома типами сервісів. Представлено метод динамічного управління ресурсами для багатоетапних систем, що використовує гнучку аналітичну модель для визначення кількості ресурсів, яку необхідно виділити для кожного етапу системи. The problem of growth of the mobile data traffic and the number of services becomes global, moreover, volume and frequency of control traffic transmitted through the network are increasing, and therefore there is a need for its effective management to ensure the quality of service required by users and optimal use of mobile network resources. In such circumstances, the load on the server that is created in the process of establishing the connection and its serving has its considerations. Wide range of modern information communication systems implies stagewise processing of incoming requests. In response to a request such systems perform a sequence of data processing actions, including, in general, number of steps. Likewise the billing system of mobile telecommunication operator involves performing succession of operations. Each service provided by the operator or service provider to the customers is charged using a distinct scheme, thus a billing system is a bottleneck in the service provisioning. This paper is devoted to the problems of multiple stage multiple resource multiple service systems. Dynamic resource provisioning in multi-stage systems is a useful technique for handling the variations seen in communication systems workloads. Virtualization technology allows to implement this approach. It is non-trivial to extend provisioning mechanisms designed for single-stage scenarios to multi-stage scenarios. Classical approaches may merely shift the bottleneck to another downstream stage. For another thing, in the networks various stages may be located on the same physical machine, and thus it is necessary to consider a common pool of the resources. An analytic model of a system would be attractive as it would be able to evaluate system characteristics under a wide range of conditions, and to be computed comparatively easily. It is also can incorporate numerical optimization techniques for system design. Moreover, there is a need in formulation and solving of the multi-criteria problem of resource allocation among different nodes with consideration for beside the failure probability other criteria – such as the cost of utilization of additional amount of resources (i.e. cost of activation of the additional server) or benefit from the certain service provisioning and so on. In this paper we argue that dynamic provisioning of multi-stage information communication systems raises new challenges not addressed by prior work on provisioning single-stage systems. An extended dynamic provisioning technique for multi-stage multi-resource multi-service systems that employs a flexible queuing model to determine how much resources to allocate to each stage of the system is proposed. The whole system is modelled as sequence of multidimensional Markov chains. Since the numerical evaluation of the closed-form failure probability in such system will become numerically intractable for a large number of traffic classes, a simpler approximate method of calculation for the state probabilities of the traffic model described is used. To evaluate the proposed technique Mathcad system is used. The effectiveness of the method is confirmed by simulation example for the system with two service types, two resource types and three stages. The proposed model is very general and can be used for analysis of different systems.uaхмарні обчисленнябагатоетапна системасистема масового обслуговуваннявиділення ресурсівcloud computingqueuing systemmulti-stage systemresource allocationArticle