Open AccessPOWER FOR PROCESSING APPLICATION FLOWS AND QUEUE SIZES IN MASS SERVICE SYSTEMS
Author(s) -
Boris Ya. Lichtzinder,
I. A. Blatov
Publication year - 2020
Publication title -
t-comm
Language(s) - English
Resource type - Journals
eISSN - 2072-8743
pISSN - 2072-8735
DOI - 10.36724/2072-8735-2020-14-9-10-16
Subject(s) - queueing theory , queue , computer science , independence (probability theory) , power (physics) , conditional independence , queue management system , mathematics , computer network , statistics , artificial intelligence , physics , quantum mechanics
The classical queuing theory studies time series processing under the assumption of sampling independence. However, the traffic of modern multiservice networks is usually strongly correlated and the methods of classical theory do not work. In this paper, we consider the cyclic process of queuing, conditional and unconditional mutual correlations. Conditional average values of queues are considered. The concept of processing power of the flow of applications in queuing systems (QS) is introduced. It is shown that the variable component of the indicated power is determined by the change in the load factor and corresponds to the conditional average size of the queue of applications in the QS.