New services and ever increasing traffic volumes require the next generation of mobile networks, e.g. 5G, to be much more flexible and scalable. The primary enabler for its flexibility is transforming network functions from proprietary hardware to software using modern virtualization technologies, paving the way of virtual network functions (VNF). Such VNFs can then be flexibly deployed on cloud data centers while traffic is routed along a chain of VNFs through software-defined networks. However, such flexibility comes with a new challenge of allocating efficient computational resources to each VNF and optimally placing them on a cluster. In this thesis, we argue that, to achieve an autonomous and efficient performance optimization method, a solid understanding of the underlying system, service chains, and upcoming traffic is required. We, therefore, conducted a series of focused studies to address the scalability and performance issues in three stages. We first introduce an automated profiling and benchmarking framework, named NFV-Inspector to measure and collect system KPIs as well as extract various insights from the system. Then, we propose systematic methods and algorithms for performance modelling and resource recommendation of cloud native network functions and evaluate them on a real 5G testbed. Finally, we design and implement a bottom-up performance simulator named PerfSim to approximate the performance of service chains based on the nodes’ performance models and user-defined scenarios. / <p>Article 5 part of thesis as manuscript, now published.</p>

The convergence of multiservice heterogeneous networks and ever increasing Internet applications, like peer to peer networking and the increased number of users and services, demand a more efficient bandwidth allocation in optical networks. In this context, new architectures and protocols are needed in conjuction with cost effective quantitative methodologies in order to provide an insight into the performance aspects of the next and future generation Internets. This thesis reports an investigation, based on efficient simulation methodologies, in order to assess existing high performance algorithms and to propose new ones. The analysis of the traffic characteristics of an OC-192 link (9953.28 Mbps) is initially conducted, a requirement due to the discovery of self-similar long-range dependent properties in network traffic, and the suitability of the GE distribution for modelling interarrival times of bursty traffic in short time scales is presented. Consequently, using a heuristic approach, the self-similar properties of the GE/G/¿ are being presented, providing a method to generate self-similar traffic that takes into consideration burstiness in small time scales. A description of the state of the art in optical networking providing a deeper insight into the current technologies, protocols and architectures in the field, which creates the motivation for more research into the promising switching technique of ¿Optical Burst Switching¿ (OBS). An investigation into the performance impact of various burst assembly strategies on an OBS edge node¿s mean buffer length is conducted. Realistic traffic characteristics are considered based on the analysis of the OC-192 backbone traffic traces. In addition the effect of burstiness in the small time scales on mean assembly time and burst size distribution is investigated. A new Dynamic OBS Offset Allocation Protocol is devised and favourable comparisons are carried out between the proposed OBS protocol and the Just Enough Time (JET) protocol, in terms of mean queue length, blocking and throughput. Finally the research focuses on simulation methodologies employed throughout the thesis using the Graphics Processing Unit (GPU) on a commercial NVidia GeForce 8800 GTX, which was initially designed for gaming computers. Parallel generators of Optical Bursts are implemented and simulated in ¿Compute Unified Device Architecture¿ (CUDA) and compared with simulations run on general-purpose CPU proving the GPU to be a cost-effective platform which can significantly speed-up calculations in order to make simulations of more complex and demanding networks easier to develop.

No

no / In this paper, a performance analysis of adaptive routing is presented for packet-switched inter-satellite link (ISL)networks, based on shortest path routing and two alternate link routing forwarding policies. The selected routing algorithm and link-cost function are evaluated for a low earth orbit satellite system, using a demographically weighted traffic flow model. Two distinct traffic flow patterns are modelled: hot spot and regional. Performance analysis, in terms of quality of service and quantity of service, is derived using specifically developed simulation software to model the ISL network, taking into account topology adaptive routing only, or topology and traffic adaptive routing.

Recently, technological developments have affected most lifestyles, especially with the growth in Internet usage. Internet applications highlight the E-commerce capabilities and applications which are now available everywhere; they receive a great number of users on a 24-7 basis because online services are easy to use, faster and cheaper to acquire. Thus E-commerce web sites have become crucial for companies to increase their revenues. This importance has identified certain effective requirements needed from the performance of these applications. In particular, if the web server is overloaded, poor performance can result, due to either a huge rate of requests being generated which are beyond the server's capacity, or due to saturation of the communication links capacity which connects the web server to the network. Recent researches consider the overload issue and explore different mechanisms for managing the performance of E-commerce applications under overload condition. This thesis proposes a formal approach in order to investigate the effects of the extreme load and the number of dropped requests on the performance of E- III commerce web servers. The proposed approach is based on the class-based priority scheme that classifies E-commerce requests into different classes. Because no single technique can solve all aspects of overload problems, this research combines several techniques including: admission control mechanism, session-based admission control, service differentiation, request scheduling and queuing model-based approach. Request classification is based on the premise that some requests (e.g. buy) are generally considered more important than others (e.g. browse or search). Moreover, this research considers the extended models from Priority Scheduling Mechanism (PSM). These models add a new parameter, such as a review model or modify the basic PSM to low priority fair model, after the discovery of ineffectiveness with low priority customers or to add new features such as portal models. The proposed model is formally specified using the π-calculus in early stage of models design and a multi-actor simulation was developed to reflect the target models as accurately as possible and is implemented as a Java-based prototype system. A formal specification that captures the essential PSM features while keeping the performance model sufficiently simple is presented. Furthermore, the simplicity of the UML bridges the gap between π-calculus and Java programming language. IV There are many metrics for measuring the performance of E-commerce web servers. This research focuses on the performance of E-commerce web servers that refer to the throughput, utilisation, average response time, dropped requests and arrival rate. A number of experiments are conducted in order to test the performance management of the proposed approaches.

De nos jours, les systèmes distribués sont caractérisés par une complexité croissante de l'architecture, des fonctionnalités et de la charge soumise. Cette complexité induit souvent une perte de la qualité de service offerte, ou une saturation des ressources, voire même l'indisponibilité des services en ligne, en particulier lorsque la charge est importante. Afin d'éviter les désagrèments causés par d'importantes charges et remplir le niveau attendu de la qualité de service, les systèmes nécessitent une auto-gestion, en optimisant par exemple un tier ou en le renforçant à travers la réplication. Cette propriété autonome requiert une modélisation des performances de ces systèmes. Visant cet objectif, nous développons un framework basé sur une méthodologie théorique et expérimentale d'identification automatique de modèle et de dimensionnement, fournissant en premier un modèle de réseau de file d'attente pour un système distribué. Ensuite, ce Modèle est utilisé au sein de notre framwork pour dimensionner le système à travers une analyse ou une simulation du réseau de file d'attente. / Modern distributed systems are characterized by a growing complexity of their architecture, functionalities and workload. This complexity, and in particular significant workloads, often lead to quality of service loss, saturation and sometimes unavailability of on-line services. To avoid troubles caused by important workloads and fulfill a given level of quality of service (such as response time), systems need to self-manage, for instance by tuning or strengthening one tier through replication. This autonomic feature requires performance modelling of systems. In this objective, we developed a framework based on a theoretical and experimental approach for automatic identification process and sizing . This framework provid a queuing model for a distributed system. Then, this model is used in our Framwork to size the system through an analysis or simulation.

Recently, technological developments have affected most lifestyles, especially with the growth in Internet usage. Internet applications highlight the E-commerce capabilities and applications which are now available everywhere; they receive a great number of users on a 24-7 basis because online services are easy to use, faster and cheaper to acquire. Thus E-commerce web sites have become crucial for companies to increase their revenues. This importance has identified certain effective requirements needed from the performance of these applications. In particular, if the web server is overloaded, poor performance can result, due to either a huge rate of requests being generated which are beyond the server¿s capacity, or due to saturation of the communication links capacity which connects the web server to the network. Recent researches consider the overload issue and explore different mechanisms for managing the performance of E-commerce applications under overload condition. This thesis proposes a formal approach in order to investigate the effects of the extreme load and the number of dropped requests on the performance of E- III commerce web servers. The proposed approach is based on the class-based priority scheme that classifies E-commerce requests into different classes. Because no single technique can solve all aspects of overload problems, this research combines several techniques including: admission control mechanism, session-based admission control, service differentiation, request scheduling and queuing model-based approach. Request classification is based on the premise that some requests (e.g. buy) are generally considered more important than others (e.g. browse or search). Moreover, this research considers the extended models from Priority Scheduling Mechanism (PSM). These models add a new parameter, such as a review model or modify the basic PSM to low priority fair model, after the discovery of ineffectiveness with low priority customers or to add new features such as portal models. The proposed model is formally specified using the ¿ -calculus in early stage of models design and a multi-actor simulation was developed to reflect the target models as accurately as possible and is implemented as a Java-based prototype system. A formal specification that captures the essential PSM features while keeping the performance model sufficiently simple is presented. Furthermore, the simplicity of the UML bridges the gap between ¿-calculus and Java programming language. IV There are many metrics for measuring the performance of E-commerce web servers. This research focuses on the performance of E-commerce web servers that refer to the throughput, utilisation, average response time, dropped requests and arrival rate. A number of experiments are conducted in order to test the performance management of the proposed approaches.

In cycling, performance models are used to investigate factors that determine performance and to optimise competition results. We present an innovative and easily applicable mathematical model describing time-resolved approaches for both the physical aspects of tractional resistance and the physiological side of propelling force generated by muscular activity and test its validity to reproduce and forecast time trials in track cycling. Six elite track cyclists completed a special preparation and two sprint time trials in an official velodrome under continuous measurement of crank force and cadence. Fatigue-free force-velocity profiles were calculated, and their fatigue-induced changes were determined by non-linear regression analysis using a monoexponential equation at a constant slope. Model parameters were calibrated based on pre-exercise performance testing and the first of the two time-trials and then used to predict the performance of the second sprint. Measured values for power output and cycling velocity were compared to the modelled data. The modelled results were highly correlated to the measured values (R2>0.99) without any difference between runs (p>0.05; d<0.1). Our mathematical model can accurately describe sprint track cycling time trial performance. It is simple enough to be used in practice yet sufficiently accurate to predict highly dynamic maximal sprint performances. It can be employed for the evaluation of completed runs, to forecast expected results with different setups, and to study various contributing factors and quantify their effect on sprint cycling performance.

Computer networks can be classified into two broad categories: wired networks and wireless networks, according to the hardware and software technologies used to interconnect the individual devices. Wired interconnection networks are hardware fabrics supporting communications between individual processors in highperformance computing systems (e.g., multi-computer systems and cluster systems). On the other hand, due to the rapid development of wireless technologies, wireless networks have emerged and become an indispensable part for people's lives. The integration of different wireless technologies is an effective approach to accommodate the increasing demand of the users to communicate with each other and access the Internet. This thesis aims to investigate the performance of wired interconnection networks and integrated wireless networks under the realistic working conditions. Traffic patterns have a significant impact on network performance. A number of recent measurement studies have convincingly demonstrated that the traffic generated by many real-world applications in communication networks exhibits bursty arrival nature and the message destinations are non-uniformly distributed. Analytical models for the performance evaluation of wired interconnection networks and integrated wireless networks have been widely reported. However, most of these models are developed under the simplified assumption of non-bursty Poisson process with uniformly distributed message destinations. To fill this gap, this thesis first presents an analytical model to investigate the performance of wired interconnection networks in multi-computer systems. Secondly, the analytical models for wired interconnection networks in multi-cluster systems are developed. Finally, this thesis proposes analytical models to evaluate the end-to-end delay and throughput of integrated wireless local area networks and wireless mesh networks. These models are derived when the networks are subject to bursty traffic with non-uniformly distributed message destinations which can capture the burstiness of real-world network traffic in the both temporal domain and spatial domain. Extensive simulation experiments are conducted to validate the accuracy of the analytical models. The models are then used as practical and cost-effective tools to investigate the performance of heterogeneous wired or wireless networks under the traffic patterns exhibited by real-world applications.

Active Queue Management (AQM) schemes are used for ensuring the Quality of Service (QoS) in telecommunication networks. However, they are sensitive to parameter settings and have weaknesses in detecting and controlling congestion under dynamically changing network situations. Another drawback for the AQM algorithms is that they have been applied only on the Markovian models which are considered as Short Range Dependent (SRD) traffic models. However, traffic measurements from communication networks have shown that network traffic can exhibit self-similar as well as Long Range Dependent (LRD) properties. Therefore, it is important to design new algorithms not only to control congestion but also to have the ability to predict the onset of congestion within a network. An aim of this research is to devise some new congestion control methods for communication networks that make use of various traffic characteristics, such as LRD, which has not previously been employed in congestion control methods currently used in the Internet. A queueing model with a number of ON/OFF sources has been used and this incorporates a novel congestion prediction algorithm for AQM. The simulation results have shown that applying the algorithm can provide better performance than an equivalent system without the prediction. Modifying the algorithm by the inclusion of a sliding window mechanism has been shown to further improve the performance in terms of controlling the total number of packets within the system and improving the throughput. Also considered is the important problem of maintaining QoS constraints, such as mean delay, which is crucially important in providing satisfactory transmission of real-time services over multi-service networks like the Internet and which were not originally designed for this purpose. An algorithm has been developed to provide a control strategy that operates on a buffer which incorporates a moveable threshold. The algorithm has been developed to control the mean delay by dynamically adjusting the threshold, which, in turn, controls the effective arrival rate by randomly dropping packets. This work has been carried out using a mixture of computer simulation and analytical modelling. The performance of the new methods that have.