Solving Problem of Optimal Placement Information Resources on the Nodes of a Distributed Information Processing System with a Self-Similar Input Flow of Applications

Abstract

This article presents the results of numerical experiments to solve the problem of optimal placement of information resources on the nodes of a distributed information processing system (DIPS) by the criterion of the minimum average response time of the system to user requests, implemented on the basis of the architecture "file-server", two-level and three-level architecture "client-server" with self-similar input traffic applications. Approximate methods of stochastic Queuing networks research (decomposition approximation method and mean value analysis method) and simulation method were used as modeling methods. The obtained results of numerical experiments are compared with the results of solving the same problems using the exact method of analysis of exponential Queuing networks. By results of the comparative analysis of the received results conclusions are drawn.