George D. Stamoulis

Application of the many sources asymptotic and effective bandwidths to traffic engineering

Accurate yet simple methods for traffic engineering are important for efficient management of resources in broadband networks. The goal of this paper is to apply and evaluate large deviation techniques for traffic engineering. In particular, we employ the recently developed theory of effective bandwidths, where the effective bandwidth depends not only on the statistical characteristics of the traffic stream, but also on a link’s operating point through two parameters, the space and time parameters, which can be computed using the many sources asymptotic. We show that this effective bandwidth definition can accurately quantify resource usage. Furthermore, we estimate and interpret values of the space and time parameters for various mixes of real traffic demonstrating how these values can be used to clarify the effects on the link performance of the time scales of traffic burstiness, of the link resources (capacity and buffer), and of traffic control mechanisms such as traffic shaping. Our experiments involve a large set of MPEG‐1 compressed video and Internet Wide Area Network (WAN) traces, as well as modeled voice traffic.

Integration of pricing and flow control for available bit rate services in ATM networks

We present a theoretically justified pricing scheme for ABR services which utilizes mechanisms provided by rate-based flow control as defined by the ATM Forum. As a result, the scheme imposes no additional communication overhead, while the added complexity at the switches and end-systems is minimal. Our approach complements the ABR rate-based flow control and leads to economically efficient utilization of network resources. According to the scheme, a connection is charged based on the sum of the price per unit of bandwidth on all links along its route. Prices depend on the demand for bandwidth and are adjusted in a decentralized and iterative manner. Simulation results show that prices converge reasonably fast and do not have a negative effect on the convergence properties of flow control.

Efficient agent-based negotiation for telecommunications services

We investigate the usage and functionality of an intelligent software agent negotiating for telecommunications services on behalf of the user. The objective of this agent is to select the best of the service combinations offered by multiple service retailers, taking into account both user preferences as encoded in the user profile and the corresponding charges given by the retailers. The agent analyzed is based on a simple yet informative model for on-line calculation of the utility of the user for the various service combinations, and on an algorithm for fast and efficient negotiation, which is not based on any assumptions regarding the structure of retailer tariffs. We assess our approach both analytically and experimentally, and we argue that the negotiation algorithm is either optimal or close to this for a variety of interesting cases. Although our approach is illustrated by dealing with a particular news-on-demand service, it is broadly applicable. We also argue that it pertains to both Internet and ATM network technologies.

Application and evaluation of large deviation techniques for traffic engineering in broadband networks

Accurate yet simple methods for traffic engineering are important for efficient dimensioning of broadband networks. The goal of this paper is to apply and evaluate large deviation techniques for traffic engineering. In particular, we employ the recently developed theory of effective bandwidths, where the effective bandwidth depends not only on the statistical characteristics of the traffic stream, but also on a links operating point through two parameters, the space and time parameters, which are computed using the many sources asymptotic. We show that this effective bandwidth definition can accurately quantify resource usage. Furthermore, we estimate and interpret values of the space and time parameters for various mixes of real traffic demonstrating how these values can be used to clarify the effects on the link performance of the time scales of burstiness of the traffic input, of the link parameters (capacity and buffer), and of traffic control mechanisms, such as traffic shaping. Our approach relies on off-line analysis of traffic traces, the granularity of which is determined by the time parameter of the link, and our experiments involve a large set of MPEG-1 compressed video and Internet Wide Area Network (WAN) traces, as well as modeled voice traffic.

Efficient Agent-Based Selection of DiffServ SLAs over MPLS Networks Within the ASP Service Model

The demand for QoS provisioning support over Internet grows continuously. One of the factors contributing to this demand is the increasing penetration of Application Service Providers (ASPs) to the market. This necessitates the development of mechanisms for the efficient realization of Service Level Agreements (SLA). In this paper, we develop and evaluate an approach for efficient SLA selection and implementation (support, policing/shaping, and charging) in a DiffServ-over-MPLS network domain. We describe how this approach is applied in a realistic service provision scenario based on the ASP service model. A negotiation process between a user and a network provider is introduced; thus the user can choose from the alternative options for allocation of resources the one that better matches his needs. For the purposes of negotiation, we develop an appropriate utility model that expresses user preferences in a simple yet informative way. Furthermore, we discuss the implementation of our approach in a small-scale experimental DiffServ-over-MPLS network, for the case of a simple scenario of ASP services provision. We also assess the economic efficiency of our approach by means of simulation experiments, the results of which advocate that our approach is incentive compatible, in the sense that individual optimization by each user (in SLA selection) also leads to improved social welfare. Our approach is quite general and can be combined with several policies for network management, or as a complement to the traffic engineering procedures.

Use of agent technology in service and retailer selection in a personal mobility context

This paper reports on work and key results of the ACTS Project MONTAGE that aims to exploit agent technology in support of personal mobility. In particular, agents are implemented to enable selection on behalf of users of the most beneficial service offer among those by multiple retailers, on the basis of user preferences encoded in the user profile, and market offerings made by retailers. A framework for agent intelligence to support service and retailer selection is proposed and justified. Enterprise, analysis and computational models of the envisaged personal mobility context that offers the capability of agent supported service and retailer selection, are provided.

QoS Management by Means of ApplicationControl

The complexity and sophistication of integratedbroadband networks call for a broad repertoire ofmanagement methods. Foreseen intensive competition ofthe communications market, and the upgrade ofend-customers satisfaction as the focal point of anymanagement policy, raise a very important problem: howto utilize efficiently the resources of the networkwhile maintaining the Quality of Service (QoS) provided to end-users within a pre-negotiated range. Inthis paper, a QoS management architecture for multimediaapplications based on an application control scheme isproposed. Adhering to the Total Quality Management (TQM) paradigm, we let management span over theentire communication stack, the entire network(end-to-end) and the entire call duration. Applicationcontrol scheme aims to meet pre-specified QoS goals by adjusting the traffic produced by theend-user applications. The adjustment technique is basedon the notion that each multimedia session is allowed tooperate on a set of discrete application quality levels. A method for monitoring and assessingthe end-user perceived quality is presented. Themathematical problem of optimal resources allocation andan associated heuristic are specified. Finally, a specific instantiation of the introduced QoSmanagement architecture on a real testbed is describedand experimentation results are presented andevaluated.

Network control and usage-based charging: is charging for volume adequate?

There has been extensive research on the applierdion of the efFective bandwidth concept for quantifying resource usage in order to create simple yet eRective usag~based charging schemes with desirable incentive properties. This research has shown that simple charging schemes which involve two measurements, time and volume, can serve their purpose well by producing adequate approximations of the effective usage of a lmrsty trailic stream. An issne that has not been addressed in detail is the relation between the type of admission control meehanism the network uses and the definition of efective usage. b particular, the above charging schemes assume “static” Connection Admission Control (C.4C). Ii contrast to static C..4C, “dynamic” C.4C strategies utilize on-line measurements of the actual load, hence can achieve mnrihhigher utilization. We argue that under such dynamic strategies the efective usage concept mnst be redefined, and that when control actions occur at faster time scales than the burstiness of the sources, the eifective usage approaches the mean rate of the sonrces. In addition to jnstifjing the potential use of simple ~olnrm+based tarii%, the above justifies the deployment of sophisticated dynamic admission control mechanism s, since these result in more competitive prices.

Grid Business Models

A business model (BM) establishes a framework for the transformation of economic inputs (e.g. resources and technological knowhow) into economic outputs (e.g. goods and services) required by customers in a market (Chesbrough and Rosenbloom 2002). In simpler terms, a business model describes the way the business expects to make money by interacting with customers and other players in the market. Such a model can also be thought of as a mediator between technology development and value creation. The ultimate role of the business model is to ensure that the technological core delivers value to the customer. In order to achieve this, a number of factors and functions must be analysed and specified, such as the value proposition of the new product, the target market, the potential value chain for the delivery of the product or service, an estimation of the cost-structure, and profit potential (Peterovic et al. 2001, Weill and Vitale 2001). A well articulated BM is the foundation of the company’s business plan. The business plan serves as a decision-support tool and includes the additional level of detail that needs to be identified and proved (as well as can be prior to execution) in order for the business to attract money from potential investors. It specifies measurable goals, the reasons why they are believed to be attainable, and the plan for reaching those goals (Siegel et al. 1993, Wikipedia 2009a). It may also include background information about the company and a marketing plan. As it becomes apparent, both are to a great extent related and equally important; without a good business model, a business plan cannot be brought to effect and vice versa.

Efficient adaptation to dynamic pricing communicated by ECN marks: Scenarios for experimental assessment

We present our research activities and initial results, regarding both simulation and actual test-bed implementation, that aim to demonstrate the feasibility and to evaluate the advantages and performance of a market-based management scheme based on a simple feedback mechanism that informs users of the congestion their traffic is experiencing. The feedback is based on Explicit Congestion Notification marking. The aforementioned objectives are investigated using two service provisioning scenarios. The first scenario involves an application provider that offers discrete Quality of Service (QoS) classes at different prices, whereas the second scenarios offers a wider range of QoS, price pairs. The experiments we report refer to the implementation of a modified version of the TCP algorithm, which provides service differentiation based on a senders willing-to-pay, and of a packet marking scheme that provides early warnings of incipient congestion.