A Technique to Solve Packing Problem for the Management of Perspective Development of Communication Network Nodes

  • PDF / 166,700 Bytes
  • 10 Pages / 594 x 792 pts Page_size
  • 32 Downloads / 185 Views

DOWNLOAD

REPORT


A TECHNIQUE TO SOLVE PACKING PROBLEM FOR THE MANAGEMENT OF PERSPECTIVE DEVELOPMENT OF COMMUNICATION NETWORK NODES A. N. Trofimchyka† and V. A. Vasyanin a‡

UDC 519.854.6:004.023

Abstract. The authors propose a technique to solve a packing problem for obtaining the initial data necessary to construct mathematical models of perspective development of nodes of multicommodity communication networks with discrete flows. The results of numerical modeling of the solution for a specific example for five stages of network development in forecasting an increase in the total flows at the nodes are given. It is shown that if network nodes are overloaded because of short-term flow oscillations or failure of several nodes, centralized solution of the optimization package problem for the whole network can always be used for immediate flow redistribution. Keywords: multicommodity networks, discrete small-batch flows, model of perspective development. INTRODUCTION The major problems relating to the software of automated control system for processing and distribution of small-batch flows in a multicommodity hierarchical network are prediction and optimization of step-by-step development of its infrastructure [1]. It is necessary to consider all possibilities of qualitative organizational and technological enhancement of the network under constraints on capital investments and their disbursement. In essence, the solution of these problems determines putting into operation of industrial capacities, fixed assets of network elements (nodal enterprises and communications of transportation network, nodes and communication channels of data transmission network) and labor force at development stages. As a rule, long-term development plans are made for several years (three to ten), one year being taken as a planning stage (planning time increments). The results of solution of such large aggregated problems, the distribution of capital investments and other material resources with respect to implementation stages, should (or can) be used to construct more detailed models at each stage. For example, for transport networks it is necessary to divide the total amount of capital investments at the stage into acquisition, modernization, and commissioning of sorting equipment (additional sorting lines), equipment for loading and discharging at nodal enterprises, vehicles for the available and new transportation routes. For data transmission networks, it is necessary to divide financial resources at the stage into modernization and re-equipment between nodes (commutators, routers) and communication channels. In each specific case, choosing the duration of the forecasting stage is a separate problem and depends on thespeed of implementation and mastering of new equipment and information technologies in network infrastructures of various economic branches. The operation stability of network structures throughout planning stages, when loads vary at nodes and network communication channels, failures occur and random factors influence, should be ensured