Аналітичні моделі багатопріоритетних хмарних дата-ценрів зі змішаною дисципліною надання послуг з урахуванням особливостей функціонування та можливих відмов

Abstract

One of the main indicators of the effectiveness of cloud data centers are indicators based on the assessment of temporal characteristics. Violation of permissible time constraints, for example, the response time of cloud data centers, has a negative impact on the efficiency of solving user target tasks. This is of particular importance for real-time systems and, above all, for special information systems built using private clouds. Analytical models of cloud data centers are developed as queuing systems with a mixed discipline of resource allocation. The models take into account failures and various features of functioning and have arbitrary distribution laws for some stochastic processes. For each model, analytical expressions for computing time characteristics have been obtained. Such models for the cloud data centers are used for the first time. General description of the models is as follows. At the input of the cloud date - center, which implements a mixed service discipline (with relative and absolute priorities), receives N Poisson flows of requests for resources with corresponding N priorities. Duration of service for applications of various flows have their own arbitrary distribution laws. An application with relative priority, whose service is interrupted by applications with absolute priority, returns to the queue. Two disciplines for the renewal of service A and B are considered. Within the same priority, applications are serviced in the order of receipt. Cloud the date-center refuses to provide resources (for example, fails) under the Poisson law, and is restored under an arbitrary law. During the recovery period, elements of adaptation to failures are used: applications of some flows to the queue are accepted and accumulated, while others are not accepted (the discipline of the queue replenishment, I and II, respectively). The denial of service device can occur both during its free state and during the application processing. Two disciplines of resumption of service after restoration C and D have been considered. An interrupted application is processed from the point of its interruption. The combination of disciplines — resumption of service and queue replenishment — allows us to consider independent models of various types of systems that have the respective designation. Different features of functioning consist of various combinations of disciplines A, B, C, D, I and II, for which analytical formulas for calculating time characteristics of different types of systems have been obtained.