Orientador : Helio Waldman / Tese (doutorado) - Universidade Estadual de Campinas, Faculdade de Engenharia Eletrica e de Computação / Made available in DSpace on 2018-08-04T03:23:02Z (GMT). No. of bitstreams: 1 AlmeidaJunior_RaulCamelodeAndrade_D.pdf: 6394727 bytes, checksum: 56d1667ed2e31b23492071088dcf7edf (MD5) Previous issue date: 2004 / Resumo: A combinação da comutação óptica de pacotes com a multiplexação por divisão de comprimento de onda (WDM) surge como um cenário promissor para as redes ópticas futuras. Um dos requisitos de projeto dessas redes é procurar minimizar a perda de pacotes quando há contendas pelos mesmos recursos. Uma alternativa é a utilização de armazenadores de pacotes ópticos, que fazem uso do domínio temporal para a resolução de contenção. Este trabalho propõe uma modelagem analítica para o desempenho desses dispositivos em redes ópticas assíncronas comutadas por pacote. Devido à inexistência de RAMs ópticas, os armazenadores de pacotes ópticos têm sido atualmente configurados por um conjunto de linhas de atraso capaz de prover um número finito de atrasos fixos. Considerando tal peculiaridade, uma contribuição adicional deste trabalho é a proposição e modelagem de duas políticas capazes de prover diferenciação de serviço em armazenadores de pacotes ópticos. Por fim, objetivando realizar um estudo comparativo entre três possíveis domínios de resolução de contenção, também são abordados os domínios espacial e do comprimento de onda, além da combinação entre eles e entre os domínios temporal e do comprimento de onda / Abstract: The combination of optical packet switching with wavelength division multiplexing (WDM) has emerged as a promising framework for future optical networks. One of the design requirements of these networks is to minimize the packet loss when there are contentions for the same resources. One possibility is to use optical buffers, which exploit the time domain for contention resolution. This work proposes an analytical modeling for the performance of such devices in asynchronous optical packet-switched networks. Due to the non-existence of optical RAMs, optical buffers have been designed using delay line banks that can provide a finite number of fixed delays. Considering such peculiarity, an additional contribution of this work is the proposal and modeling of two policies that can provide service differentiation in optical buffers. Lastly, in order to make a comparative study among three possible contention resolution domains, the space and wavelength domains are also analyzed, as well as their combinations and the combination between the time and wavelength domains / Doutorado / Telecomunicações e Telemática / Doutor em Engenharia Elétrica

M.Ing. / The Internet is evolving into a global communication infrastructure that is expected to support an overabundance of new applications such as IP telephony, interactive TV, and e-commerce. The existing best effort service is no longer sufficient. It is not enough to provide differentiated services and to meet QoS requirements of these different traffic types. As a result, there is an urgent need to provide more services that are powerful such as guaranteed services, flow protection etc, merged in one IP network, referred to as Integrated Services Packet-Switching Network (ISPN) in this thesis. To provide these services, QoS aware network architectures are required to implement the services. This dissertation presents a survey on two network architectures: Fair Queuing (FQ) and Scalable Core (SCORE), which attempt to provide QoS solutions in ISPN. We theoretically analyse scheduling as an important element in providing QoS in these architectures. The important thread in scheduling is performance and implementation complexity. SCORE based scheduling have less implementation complexity but cannot exactly match the high performance of FQ solutions, which suffer implementation complexity. The contribution of this work is a feedback protocol that minimises congestion in SCORE scheduling scheme called Core stateless fair queuing (CSFQ). The flow rates are adjusted by sending rate signal to a transmitting node from a receiving node, to adjust ill-behaved flow rate during congestion to a fair share rate of receiving node. We use CSFQ based theoretical analysis and simulations to demonstrate the performance of the feedback protocol.

No description available.

A system composed of unreliable components can experience different levels of performance as its configuration changes due to failures and repairs. One approach used to measure overall system performance is to weight the level of performance measured for each system state by the probability that the system is in that state and then sum across all system states. Many performance measures have a transient behavior following a change in the state of the system. Because of the difficulty associated with transient analysis, the system is often assumed to be in steady state when measuring the performance for each system state. When this approach is used to analyze packet-switched communication networks, which consist of highly reliable high-speed links and switching nodes, it is argued that the steady-state assumption is justified on the basis of the large difference in rates of traffic-related events, such as call completions and packet transmissions, compared to component-related events, such as failures and repairs. To investigate the validity of this assumption, we define lower bounds for the length of the transient phase fol1owing link failures and repairs. For both cases, we obtain a distribution for the length of the lower bound. The transient phase is significant when its length exceeds a given fraction of the time until the next change in network state. Using the distributions for these lengths, we derive an expression for the probability that the transient phase is significant in terms of the amount of traffic on the link and the ratio of the rates for traffic-related events and network state changes. These results show that the difference in rates between traffic-related events and component related events is not enough by itself to justify the steady-state assumption. The amount of traffic carried on the link and the size of the network must also be considered. These results indicate some situations where the steady-state assumption is inappropriate. We also obtain sufficient conditions for transient-phase significance following link failures. Although these results do not indicate when it is safe to use the steady-state assumption, they provide a measure of the risk associated with using it. / Ph. D.

This research investigates the implementation and performance of packet video transfer over local area networks. A network architecture is defined for packet video such that most of the processing is performed by the higher layers of the Open Systems Interconnection (OSI) reference model, while the lower layers provide real-time services. Implementation methods are discussed for coding schemes, including data compression, the network interface unit, and the underlying local area network (LAN), Ethernet or the Fiber Distributed Data Interface (FDDI). Performance evaluation is presented using simulation results and analyses for different video sources, implementation models, and LAN s. The simulation experiments are performed for systems where video images are retrieved from databases at one or more servers and delivered over the local area network. / Master of Science

No description available.

Thesis (Elec.E.)--Massachusetts Institute of Technology, Dept. of Electrical Engineering and Computer Science, 1981. / Includes bibliographical references. / by Howard Paul Hayden. / Thesis (Elec.E.)--Massachusetts Institute of Technology, Dept. of Electrical Engineering and Computer Science, 1981.

Application performance continues to be an issue even with increased Internet bandwidth. There are many reasons for poor application performance including unpredictable network conditions, long round trip times, inadequate transmission mechanisms, or less than optimal application designs. In this work, we propose to exploit flow relationships as a general means to improve Internet application performance. We define a relationship to exist between two flows if the flows exhibit temporal proximity within the same scope, where a scope may either be between two hosts or between two clusters of hosts. Temporal proximity can either be in parallel or near-term sequential. As part of this work, we first observe that flow relationships are plentiful and they can be exploited to improve application performance. Second, we establish a framework on possible techniques to exploit flow relationships. In this framework, we summarize the improvements that can be brought by these techniques into several types and also use a taxonomy to break Internet applications into different categories based on their traffic characteristics and performance concerns. This approach allows us to investigate how a technique helps a group of applications rather than a particular one. Finally, we investigate several specific techniques under the framework and use them to illustrate how flow relationships are exploited to achieve a variety of improvements. We propose and evaluate a list of techniques including piggybacking related domain names, data piggybacking, enhanced TCP ACKs, packet aggregation, and critical packet piggybacking. We use them as examples to show how particular flow relationships can be used to improve applications in different ways such as reducing round trips, providing better quality of information, reducing the total number of packets, and avoiding timeouts. Results show that the technique of piggybacking related domain names can significantly reduce local cache misses and also reduce the same number of domain name messages. The data piggybacking technique can provide packet-efficient throughput in the reverse direction of a TCP connection without sacrificing forward throughput. The enhanced ACK approach provides more detailed and complete information about the state of the forward direction that could be used by a TCP implementation to obtain better throughput under different network conditions. Results for packet aggregation show only a marginal gain of packet savings due to the current traffic patterns. Finally, results for critical packet piggybacking demonstrate a big potential in using related flows to send duplicate copies to protect performance-critical packets from loss.

Thesis (Sc.D.)--Massachusetts Institute of Technology, Dept. of Electrical Engineering and Computer Science, 1981. / MICROFICHE COPY AVAILABLE IN ARCHIVES AND ENGINEERING. / Bibliography: leaves 104-106. / by Oliver Chukwudi Ibe. / Sc.D.

Sun Weng Hong. / Thesis submitted in: October 2002. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2003. / Includes bibliographical references (leaves 116-121). / Abstracts in English and Chinese. / Chapter 1. --- Introduction --- p.7 / Chapter 2. --- Ant Colony Optimization (ACO) --- p.9 / Chapter 2.1 --- ACO vs. Traditional Routing --- p.10 / Chapter 2.1.1 --- Routing information --- p.10 / Chapter 2.1.2 --- Routing overhead --- p.12 / Chapter 2.1.3 --- Adaptivity and Stagnation --- p.14 / Chapter 2.2 --- Approaches to Mitigate Stagnation --- p.15 / Chapter 2.2.1 --- Pheromone control --- p.15 / Chapter 2.2.1.1 --- Evaporation: --- p.15 / Chapter 2.2.1.2 --- Aging: --- p.16 / Chapter 2.2.1.3 --- Limiting and smoothing pheromone: --- p.17 / Chapter 2.2.2 --- Pheromone-Heuristic Control --- p.18 / Chapter 2.2.3 --- Privileged Pheromone Laying --- p.19 / Chapter 2.2.4 --- Critique and Comparison --- p.21 / Chapter 2.2.4.1 --- Aging --- p.22 / Chapter 2.2.4.2 --- Limiting pheromone --- p.22 / Chapter 2.2.4.3 --- Pheromone smoothing --- p.23 / Chapter 2.2.4.4 --- Evaporation --- p.25 / Chapter 2.2.4.5 --- Privileged Pheromone Laying --- p.25 / Chapter 2.2.4.6 --- Pheromone-heuristic control --- p.26 / Chapter 2.3 --- ACO in Routing and Load Balancing --- p.27 / Chapter 2.3.1 --- Ant-based Control and Its Ramifications --- p.27 / Chapter 2.3.2 --- AntNet and Its Extensions --- p.35 / Chapter 2.3.3 --- ASGA and SynthECA --- p.40 / Chapter 3. --- Multiple Ant Colony Optimization (MACO) --- p.45 / Chapter 4. --- MACO vs. ACO --- p.51 / Chapter 4.1 --- Analysis of MACO vs. ACO --- p.53 / Chapter 5. --- Applying MACO in Load Balancing --- p.89 / Chapter 5.1 --- Applying MACO in Load-balancing --- p.89 / Chapter 5.2 --- Problem Formulation --- p.91 / Chapter 5.3 --- Types of ant in MACO --- p.93 / Chapter 5.3.1 --- Allocator. --- p.94 / Chapter 5.3.2 --- Destagnator. --- p.95 / Chapter 5.3.3 --- Deallocator. --- p.100 / Chapter 5.4 --- Global Algorithm --- p.100 / Chapter 5.5 --- Discussion of the number of ant colonies --- p.103 / Chapter 6. --- Experimental Results --- p.105 / Chapter 7. --- Conclusion --- p.114 / Chapter 8. --- References --- p.116 / Appendix A. Ants in MACO --- p.122 / Appendix B. Ants in SACO. --- p.123