PERFORMANCE REQUIRED FOR COMMON-USE COMPONENTS OF COMPUTER NETWORKS

Sometimes, in practice, simple solutions of quick preliminary estimation of basic characteristics of a computer network are needed.In this aim, the backbone subnet and server set of wide area computer networks are examined. Based on Jackson’s partitioning theorem and considering the linear...

Full description

Saved in:
Bibliographic Details
Main Author: BOLUN, Ion
Format: Article
Language:English
Published: Technical University of Moldova 2024-09-01
Series:Journal of Engineering Science (Chişinău)
Subjects:
Online Access:https://press.utm.md/index.php/jes/article/view/2024-31-3-06/06-pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Sometimes, in practice, simple solutions of quick preliminary estimation of basic characteristics of a computer network are needed.In this aim, the backbone subnet and server set of wide area computer networks are examined. Based on Jackson’s partitioning theorem and considering the linear dependence of the costs of channels, routers, and servers on their performance, a simplified analytical model for these components of the network is defined. Using this model, two optimization problems are formulated: minimizing the average response time to user requests of data processing and minimizing the summary cost of servers, channels and routers of the computer network. For both problems, analytical solutions regarding the necessary performances of channels, routers and servers are obtained. As expected, in the obtained analytical solutions, the equations for the optimization criteria of the two problems coincide, only their form being different. Calculations of performances according to these solutions are simple and can be done, for example, in MS Excel. Because the obtained in this way performances are positive real numbers, and the allowed performances of concerned entities are discrete ones, further adjustment of the solution in question, depending of the case, may be necessary. For such an adjustment, two algorithms are proposed. One of them solves the problem by reducing it to that of backpack. Another solves the problem based on the use of resource concentration rule.
ISSN:2587-3474
2587-3482