The introduction of a new technology High Speed Downlink Packet Access (HSDPA) in the Release 5 of the 3GPP specifications raises the question about its performance capabilities. HSDPA is a promising technology which gives theoretical rates up to 14.4 Mbits. The main objective of this thesis is to discuss the system level performance of HSDPAMainly the thesis exploration focuses on the Packet Scheduler because it is the central entity of the HSDPA design. Due to its function, the Packet Scheduler has a direct impact on the HSDPA system performance. Similarly, it also determines the end user performance, and more specifically the relative performance between the users in the cell.The thesis analyzes several Packet Scheduling algorithms that can optimize the trade-off between system capacity and end user performance for the traffic classes targeted in this thesis.The performance evaluation of the algorithms in the HSDPA system are carried out under computer aided simulations that are assessed under realistic conditions to predict the results as precise on the algorithms efficiency. The simulation of the HSDPA system and the algorithms are coded in C/C++ language

none

Asynchronous Transfer Mode (ATM) network is being deployed in carrier backbone. ATM can transmit a wide variety of traffic, such as video, voice, and data. Available Bit Rate (ABR) service is one of six ATM services, which is now under intensive research for its closed loop feedback control feature. ABR service supports two types of connections: unicast and multicast. There are also three types of multicast connections: point-to-multipoint, multipoint-to-point, and multipoint-to-multipoint. Multipoint communication is the exchange of information among multiple senders and multiple receivers, forming a multicast group. Examples of multicast applications include audio and video conferencing, video on demand, tele-metering, distributed games, and data distribution applications. In this dissertation, we focus on queuing and packet scheduling management for multipoint-to-point ABR connections. Although there are so many proposed fairness definitions for all ABR sources in a multipoint-to-point connection, there are still problems about queue lengths, queuing delays, and throughputs, when ABR sources are with variable-length packets. From the nature of VC-merge scheme on merged points in a multipoint-to-point connection, merged switches cannot transmit cell-stream of a packet out until the packet is completely and totally queued. If there is no complete packets queued, the switch can then choose an incomplete packet for cut-through forwarding for efficiency. Therefore, if the switch chooses a long packet from a branch that has smaller cell input rate, for cut-through forwarding, the throughput of output ports will experience severe oscillations. At the same time, ABR queue lengths will be also occupied with severe growth, and ABR cells will be experienced long queuing delays. We proposed a scheme, named MWTF (Minimum Waiting Time First), which is architecture-independent of any rate allocation schemes and fairness definitions, to resolve the problems by providing length of each packet to merged switches. Thereby the scheduler can choose an appropriate incomplete packet for cut-through forwarding, by selecting the packet that has the smallest packet waiting time. Simulation results show that merged switch has good performances. Throughput will be no severe oscillations and will be getting smoother. Also cells have smaller and smoother queuing delays in average, and the switches have much smaller queue lengths and smoother variations.

Traditional TCP when applied in wireless networks may encounter two limitations. The first limitation is the higher bit error rate (BER) due to noise, fading, and multipath interference. Because traditional TCP is designed for wired and reliable networks, packet loss is mainly caused by network congestions. As a result, TCP may decrease congestion window inappropriately upon detecting a packet loss. The second limitation is about the packet scheduling, which mostly does not consider wireless characteristics. In this Thesis, we propose a local retransmission mechanism to improve TCP throughput for wireless networks with higher BER. In addition, we measure the packet loss ratio (PLR) to adjust the queue weight such that the available bandwidth for each queue can be changed accordingly. In our mechanism, the queue length is used to determine whether there is a congestion in wireless networks. When the queue length exceeds a threshold, it indicates that the wireless networks may have congestion very likely. We not only propose the dynamic weight-adjustment mechanism, but also solve the packet out-of-sequence problem, which results form when a TCP flow changes to a new queue. For the purpose of demonstration, we implement the proposed weight-adjustment mechanisms on the Linux platform. Through the measurements and discussions, we have shown that the proposed mechanisms can effectively improve the TCP throughput in wireless networks.

Resource virtualization is emerging as a technology to enable the management and sharing of hardware resources including multiple core processors and accelerators such as Digital Signal Processors (DSP), Graphics Processing Units (GPU), and Field Programmable Gate Arrays (FPGA). Accelerators present unique problems for virtualization and sharing due to their specialized architectures and interaction modes. This thesis explores and proposes solutions for the virtualized operation of high performance, quality of service (QoS) packet scheduling accelerators. It specifically concentrates on challenges to meet 10Gbps Ethernet wire speeds. The packet scheduling accelerator is realized in a FPGA and implements the Sharestreams-V architecture. ShareStreams-V implements the Dynamic Window-Constrained Scheduler (DWCS) algorithm, and virtualizes the previous ShareStreams architecture. The original ShareStreams architecture, implemented on Xilinx Virtex-I and Virtex-II FPGAs, was able to schedule 128 streams at 10Gbps Ethernet throughput for 1500-byte packets. Sharestreams-V provides both hardware and software extensions to enable a single implementation to host isolated, independent virtual schedulers. Four methods for virtualization of the packet scheduler accelerator are presented: coarse- and fine-grained temporal partitioning, spatial partitioning, and dynamic spatial partitioning. In addition to increasing the utilization of the scheduler, the decision throughput of the physical scheduler can be increased when sharing the physical scheduler across multiple virtual schedulers among multiple processes. This leads to the hypothesis for this work: Virtualization of a quality of service packet scheduler accelerator through dynamic spatial partitioning is an effective and efficient approach to the accelerator virtualization supporting scalable decision throughput across multiple processes. ShareStreams-V was synthesized targeting a Xilinx Virtex-4 FPGA. While sharing among four processes, designs that supported up to 16, 32, and 64 total streams are able to reach 10Gbps Ethernet scheduling throughput for 64-byte packets. When sharing among 32 processes, a scheduler supporting 64 total streams was able to reach the same throughput. An access API presents the virtual scheduler abstraction to individual processes in order to allocate, deallocate, update and control virtual the scheduler allocated to a process. Practically, the bottleneck for the test system is the software to hardware interface. Effective future implementations are anticipated to use a tightly-coupled host CPU to accelerator interconnect.

[Truncated abstract] The accelerating pace of convergence of communications from disparate application types onto common packet networks has made quality of service an increasingly important and problematic issue. Applications of different classes have diverse service requirements at distinct levels of importance. Also, these applications offer traffic to the network with widely variant characteristics. Yet a common network is expected at all times to meet the individual communication requirements of each flow from all of these application types. One group of applications that has particularly critical service requirements is the class of real-time applications, such as packet telephony. They require both the reproduction of a specified timing sequence at the destination, and nearly instantaneous interaction between the users at the endpoints. The associated delay limits (in terms of upper bound and variation) must be consistently met; at every point where these are violated, the network transfer becomes worthless, as the data cannot be used at all. In contrast, other types of applications may suffer appreciable deterioration in quality of service as a result of slower transfer, but the goal of the transfer can still largely be met. The goal of this thesis is to evaluate the potential effectiveness of a class of packet scheduling algorithms in meeting the specific service requirements of real-time applications in a converged network environment. Since the proposal of Weighted Fair Queueing, there have been several schedulers suggested to be capable of meeting the divergent service requirements of both real-time and other data applications. ... This simulation study also sheds light on false assumptions that can be made about the isolation produced by start-time and finish-time schedulers based on the deterministic bounds obtained. The key contributions of this work are as follows. We clearly show how the definition of the virtual time function affects both delay bounds and delay distributions for a real-time flow in a converged network, and how optimality is achieved. Despite apparent indications to the contrary from delay bounds, the simulation analysis demonstrates that start-time rate-based schedulers possess useful characteristics for real-time flows that the traditional finish-time schedulers do not. Finally, it is shown that all the virtual time rate-based schedulers considered can produce isolation problems over multiple hops in networks with high loading. It becomes apparent that the benchmark First-Come-First-Served scheduler, with spacing and call admission control at the network ingresses, is a preferred arrangement for real-time flows (although lower priority levels would also need to be implemented for dealing with other data flows).