تحديث الشبكة الضوئية الرئيسية السودانية بتعيين مواقع مراكز البيانات باستخدام البرمجة الخطية

Abstract EN

With the development of Internet technology and the expansion of its services, the need for storage and fast transfer of data to end users has increased.

Therefore, it is necessary to connect the network such that cost is minimized.

In addition, data center locations should be optimized to reduce upload and download cost.

This paper aims to find the best way to connect major Sudanese cities with optical fibers and to compare the result with the current network and subsequently propose the best way to enhance it.

It also aims to identify the ideal cities to locate data centers.

This is done based on a number of factors such as distances between cities, optical link capacity, number of links, and the traffic demand between cities.

Mixed Integer Linear Programming (MILP) has been used to develop two models: the first model that optimizes the network topology and the second model that optimizes data center locations.

The optical network is considered using Wavelength Division Multiplexing (WDM).

Major Sudanese cities with the largest population were selected to test the models, where the population was used to determine the city's demand for data.

The network topology optimization model produced the best way to connect Sudanese cities, and was found to be similar to the current optical network.

The model was then employed to propose the best way to connect the network by increasing the number of links connecting cities.

Using the data center location optimization model, the optimum cities to deploy data centers are found, when the number of data centers in the network is one to six data centers.

The optimum network topology depends on the number of nodes, the distance between them and the number of links.

Whereas the optimum data center locations depend on their number in the network and the amount of upload and download data.