Optimization-based resource allocation in communication networks

Abstract

The continuously growing number of applications competing for resources in current communication networks highlights the necessity for efficient resource allocation mechanisms to maximize user satisfaction. Optimization Theory can provide the necessary tools to develop such mechanisms that will allocate network resources optimally and fairly among users. The aim of this paper is to provide a starting point for researchers interested in applying optimization techniques in the resource allocation problem for current communication networks. To achieve that we, first, describe the fundamental optimization theory tools necessary to design optimal resource allocation algorithms. Then, we describe the Network Utility Maximization (NUM) framework, a framework that has already found numerous applications in network optimization, along with some recent advancements of the initial NUM framework. Finally, we summarize some of our recent work in the area and discuss some of the remaining research challenges towards the development of a complete optimization-based resource allocation protocol.

Introduction

Since the creation of ARPANET [1], the first packet switching communication network in 1969 that connected university laboratories, industrial and government research centers in the US, there has been a tremendous change in the extent and characteristics of communication networks, and especially the internet, the amount of data that is shared through them and the variety of applications that generate this traffic.

Recent Cisco IP traffic studies [2], [3] provide useful information and insights regarding the traffic characteristics in current communication networks. The total internet traffic currently exceeds 40 Exa-bytes per month. On the other hand, mobile traffic has seen an explosive increase in the past years. While total mobile traffic in 2008 was no more than 33 Peta-bytes per month, mobile traffic is forecasted to reach 2.1 Exa-bytes per month by the end of 2013. The reason causing this abrupt increase in the traffic in both internet and mobile networks can be justified if one looks carefully at these statistics from another perspective; that of the applications that generate the traffic. In 2008, the majority of the traffic was generated by “traditional” types of applications, such as web browsing, email and file sharing applications, that accounted for 77% of the total traffic in the internet. However, multimedia applications, such as VoIP and video streaming, dominate the traffic nowadays exceeding 57% of the total traffic in the internet. The statistics are similar in the mobile internet as well, where the video traffic alone currently accounts for two-thirds of the total mobile traffic.

This abrupt increase of the total traffic highlights the necessity for more efficient methods of sharing the available bandwidth so that users are receiving the maximum possible satisfaction and the best possible experience when using a communication network. In addition, taking into account that users are being charged by the network providers for access, the more efficient the resource allocation is, the more satisfied the users will be and consequently the more willing to continue paying the provider for the service. The heterogeneity of the provided applications also shows that all traffic does not have the same resource requirements. This strengthens the need for more sophisticated allocation methods that will be able to distinguish between different types of applications and try to allocate resources in a way that maximizes user satisfaction for each application type.

Optimization Theory can provide a powerful tool in the development of such methods for various reasons. Optimization Theory has been used successfully in many areas related to communication networks, such as optimal routing, flow control and power control. In particular, pioneering and early work that is based on optimization techniques on link capacity assignment, routing and flow control in communication networks can be found in, e.g., [4], [5], [6]. Besides these, optimization theory is also widely applied to other applications, such as chemical engineering [7], fleet management and inventory organization, since it leads to the best possible solutions for a given problem. In addition, there are techniques, such as the Langrangian Method that can lead to the development of distributed algorithms. Distributed calculation of the optimal solution is of significant importance in communication networks, which consist of numerous network nodes and traffic sources that behave independently and selfishly to achieve the maximum possible level of satisfaction using the resources of the network. Moreover, optimization theory can also be used to assure that the allocation of resources to each application will follow its Quality of Service requirements and satisfy some notion of fairness. This can be achieved by the appropriate formulation of the optimization problem and the use of specific allocation policies according to the desired type of fairness.

The purpose of this paper is twofold; first, to introduce Lagrangian Optimization as a technique to solve the resource allocation problem in communication networks, and then, to provide some samples of recent work in the area and outline some of the most important challenges that need to be tackled in the future. The rest of the paper is organized as follows. Section 2 provides a summary of the most important notions of Optimization Theory, necessary for someone to do research in the area. Then, Section 3 describes the Network Utility Maximization (NUM) framework and the most important enhancements of the initial NUM framework. Then, Section 4 provides an overview of our recent research in the area as a motivation to tackle some of the remaining open research challenges, which are summarized in Section 5.