OFS provisions bandwidth in the granularity of one wavelength. With such a coarse granularity, most applications including video download, HDTV, 3D movie, and 3D TV etc. will have very short flow sizes, in the order of seconds or even sub-second, which brings challenges to the utilization efficiency of bandwidth capacity. In this thesis we study the performance of OFS for short flows. The constraint of network resources is investigated. The effect of destination and path blocking is studied. A distributed scheduling protocol called Multi-Destination Control Protocol (MDCP) is proposed to deal with such constraint. Both single wavelength and multi-wavelength configurations are studied and characterized. Simulation results demonstrate that MDCPcan improve the OFS network throughput significantly and can be as much as eighty to one hundred percent for a single-wavelength OFS network. Even for an OFS network with four wavelengths, the throughput improvement can still approach 40%. / The Internet traffic has been growing tremendously. China Telecom predicts that the compound annual growth rate of IP traffic for the next decade is at 56% - 80% and the backbone capacity will grow by another two orders of magnitudes. Furthermore, the power consumption incurred by the next generation of huge electronic IP packet switching routers in the backbone will exceed gigawatts. In view of the grave enviromnental concerns, there is a great need for a more efficient way of transporting and switching the bits. This thesis investigates a new all-optical networking technology called optical flow switching (OFS). OFS bypasses electronic routers, and provides end-to-end transparent connections, thus taking full advantage of the enormous transmission capacity of optical networks and enjoying the extremely low error rate of transparent data transmission. The most important point about OFS is that it reduces the electrical power consumption by off-loading the huge electronic routers, which could be a major constraint for future Internet growth. Unlike many other exotic all-optical switching technologies, OFS is immediately deployable using the current optical technologies, Therefore OFS is very attractive for the next generation optical networks. / Qian, Zhengfeng. / Adviser: Kwok-wai Cheung. / Source: Dissertation Abstracts International, Volume: 73-04, Section: B, page: . / Thesis (Ph.D.)--Chinese University of Hong Kong, 2011. / Includes bibliographical references (leaves 113-118). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. [Ann Arbor, MI] : ProQuest Information and Learning, [201-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstract also in Chinese.

An increasing number of network applications use reliable transport protocols. Applications with constant data transmission recover from loss without major performance disruption, however, applications that send data sporadically, in small packets, also called thin-streams, experience frequently high latencies due to 'Bufferbloat', that reduce the application performance. Active Queue Management mechanisms were proposed to dynamically manage the queues in routers by dropping packets early and reduce these, hence reducing latency. While their deployment to the internet remains an open issue, the proper investigation into how their functioning mechanism impacts latency is the main focus of this work and research questions have been devised to investigate the AQM impact on latency. A range of AQM mechanisms has been evaluated by the research, exploring performance of the methods for latency sensitive network applications. This has explored new single queue AQM mechanisms such as Controlled Delay (CODEL) and Proportional Integral Enhanced (PIE) and Adaptive RED (ARED). The evaluation has shown great improvements in queuing latency when AQM are used over a range of network scenarios. Scheduling AQM algorithms such as FlowQueue CODEL (FQ-CODEL) isolates traffic and minimises the impact of Bufferbloat on flows. The core components of FQ-CODEL, still widely misunderstood at the time of its inception, have been explained in depth by this study and their contribution to reducing latency have been evaluated. The results show significant reductions in queuing latency for thin streams using FQ-CODEL. When TCP is used for thin streams, high application latencies can arise when there are retransmissions, for example after dropping packets by an AQM mechanism. This delay is a result of TCP's loss-based congestion control mechanism that controls sender transmission rate following packet loss. ECN, a marking sender-side improvement to TCP reduces applicationlayer latency without disrupting the overall network performance. The thesis evaluated the benefit of using ECN using a wide range of experiments. The findings show that FQ-CODEL with ECN provides a substantial reduction of application latency compared to a drop-based AQM. Moreover, this study recommends the combination of FQ-CODEL with other mechanisms, to reduce application latency. Mechanisms such as ABE, have been shown to increase aggregate throughput and reduce application latency for thin-stream applications.