DonNTU   Masters' portal

Abstract

Content

Introduction

With the growing number of services and the expansion of its presence more and more companies and organizations come to the conclusion that increasing the amount of data within their own corporate network. Creating a multi-(voice, video, data), a geographically distributed infrastructure allows you to use the full potential of modern information technology, which makes it possible to establish effective functioning of the company and streamline internal business processes.

Arguments in favor of multi-service networks:
• optimizing the bandwidth of communication channels;
• more efficient use of communication channels;
• reducing the cost of running the same infrastructure;
• flexible expansion and implementation of new information services;
• Saves resources and channel lease purchase of equipment access to the global network;
• communication reliability and security;
• A significant expansion of the scope of application of information technology to improve business performance.

Example of multi-service network:
• telephone and fax;
• digital channels at a constant rate;
• Packet with the required quality of service;
• Transfer images, video conferencing;
• TV;
• IP-telephony;
• Broadband access to the Internet;
• pair with remote local area networks;
• Creation of switched and managed virtual private networks.

1. Theme urgency

Modern telecommunication services, especially services (real-time video conferencing, VoIP) connections require a specified quality of service, which in turn puts one of the key challenges – traffic management [1]. Under the direction of traffic refers to the set of algorithmic tools implemented both hardware and software to ensure the functioning of the network in question with the required quality of service and efficient use of resources. Are engaged in routing protocols that collect information about the topology of interconnection. Recently algorithms are developed multihoming the transfer of data of one logical connection to several physical channels. Their implementation, in many cases, a more efficient use channel resources when transferring large data streams.

2. Goal and tasks of the research

The main purpose of this work is to develop an algorithm of multihoming based on the methods QoS routing.
To achieve this goal it is necessary to solve the following problems:
1. Analyze existing research and development in this area;
2. To review the principles of routing algorithms;
3. Analyze the principles of QoS-routing;
4. Based on QoS-routing algorithm to develop an algorithm multihoming;
5. Evaluate the results and effectiveness of this algorithm.

As part of the master's work is planned for the actual scientific results:
Multihoming planned to develop an algorithm that can effectively programmatically allocate network resources. It is also planned to develop a generator of self-similar traffic based on the ON / OFF sources for the study of the influence of self-similarity on the algorithm multihoming [2,3].

3. QoS-routing, algorithms of multihoming

Occupies an important place routing to provide a quality of service – QoS-routing. The task of QoS-routing is to find such a path between a pair of nodes for which the data transfer values of some parameters on the quality of service of a connection defined by the user or the provider will not be violated.
Currently, there are a lot of QoS-routing algorithms. One algorithm involves the use of metrics, built on a linear combination of the two metrics that show different channel parameters to find the shortest path. The main disadvantages are: use of a linear function, which leads to incorrect results and the operation of high computational complexity. Another algorithm is to provide a consistent calculation of the executable path. First, is the best way (the way), one of the metrics and then the algorithm checks the optimality for the other metrics. In case of default set of conditions optimization is performed by a different metric for as long as it finds a feasible path.
For the data transmission mechanism responsible transport layer protocols. Results of this level are to communicate point-to-point. EXAMPLE: TCP, UDP, SCTP [4,5,6].
Figure 2 [7] The classification of transport layer protocols. Particular attention should be paid to the specialized transport protocol SCTP (Stream Control Transmission Protocol) – Stream Control Transmission Protocol. SCTP provides transport-layer functions for most applications network TCP / IP. The basic original properties of SCTP are multi-threading and multihoming [8,9].

Principle of Multihoming

Figure 1 – Principle of multihoming
(animation: 6 frames, 1 s delay, size 85 kB, an unlimited number of repetitions, made with gifovina.ru)

Figure 2 – Classification of transport layer protocols.

During the study, a model was developed using the concept of network multihoming. To simulate traffic NASA was taken one month [10]. Multihoming developed algorithm is shown in Figure 3. In the study, was used several modifications of the algorithm. In addition to a comprehensive optimization criterion applied by individual characteristics of the channel. Based on the results, the use of complex criterion was the most effective.

Figure 3 – Algorithm of Multihoming.

 

Conclusion

Multithreading and multihoming significantly increases the performance and reliability of telecommunication networks. However, applying this concept is not always appropriate. For example the work of real-time services in the application multihoming may be incorrect or impossible because of the different characteristics or dynamically changing channels, more profitable use of multithreading for high volumes of traffic-independent real-time. There is also a number of disadvantages: If all node connections are available all the time chart during a session, but need not be, it is not optimal from the economic point of view, on the other hand, to connect to a network node for a certain time required to for which the data loss will occur. The solution to this problem is to use a long-term prediction of the behavior of network traffic, for the timely preparation of the necessary communication channel. With this combination of techniques, advantages multihoming concept will be used in full.

This master's work is not completed yet. Final completion: December 2013. The full text of the work and materials on the topic can be obtained from the author or his head after this date.

References

    1. Гришаева, А.Д., Алтухов, Д.С., Дегтяренко, И.В. Применение механизма фаззи-логики для распределения потоков трафика в гетерогенной мультиоператорской среде // Автоматизація технологічних об’єктів та процесів. Пошук молодих. Збірник наукових праць ХІI науково-технічної конференції аспірантів та студентів в м. Донецьку 17-20 квітня 2012 р. - Донецьк, ДонНТУ, 2012. – с.20-22.
    2. Montavont, N., et al.: Analysis of Multihoming in Mobile IPv6. draft-ietf-monami6-mipv6-analysis-01 (June 26, 2006).
    3. Hiroyuki Koga, Hiroaki Haraguchi, Katsuyoshi Iida, and Yuji Oie. A Framework for Network Media Optimization in Multihomed QoS Networks/ National Institute of Information and Communications Technology, Japan. – 2009.
    4. Дегтяренко І.В., Абраменко О.О., Чекунков О.С. Прогностичне керування навантаженням на сервери з використанням нейромережі/ Дегтяренко І.В., Абраменко О.О., Чекунков О.С.// Наукові праці Донецького національного технічного університету. / Збірник наукових праць ДонНТУ. Серія: Обчислювальна техніка та автоматизація. Випуск: 23 (201). — Донецьк, ДонНТУ, 2012. — C. 95–102.
    5. Гаськова И. А. Использование пакета OPNET IT GURU EDITION при проектировании мультисервисных телекоммуникационных сетей / Гаськова И. А.// Материалы Научно-технической конференции «Проблемы телекоммуникаций». Сборник тез. К.: НТУУ "КПИ", 2010. —188 с.
    6. Олифер В.Г., Олифер Н.А. Компьютерные сети. Принципы, технологии, протоколы : Учебник для вузов. 4-е изд. – СПб.: Питер, 2010. – c.355-378.
    7. Программа сетевой академии Cisco CCNA 1 и 2. Вспомогательное руководство, 3-е изд., с испр.: Пер. с англ. – М.:Издательский дом «Вильямс», 2008. – с.768-801.
    8. Протокол маршрутизации EIGRP [Електронный ресурс] // NETCONFIG.ORG. - Режим доступа : http://www.netconfig.org/category/routing/eigrp/.
    9. Кучерявый Е.А. Управление трафиком и качеством обслуживания в сети Интернет. – СПб.: Наука и Техника, 2004. – c.247-254.
    10. Two traces contain two month's worth of all HTTP requests to the NASA Kennedy Space Center WWW server in Florida [электронный ресурс]. – Режим доступа:http://ita.ee.lbl.gov/html/contrib/NASA-HTTP.html.