Design and Analysis Techniques for Future High-Speed Networks

与传统的数据网络不同，未来的宽带- isdn （BISDN）广域网将需要承载从突发的可变速率源（如语音和可变速率编码视频）到平滑的恒定比特率源的各种通信量。此外，这些网络必须在为这些流量类提供有保证的性能或服务质量（QOS）的同时这样做。因此，在这种网络中表征性能的问题特别重要，因为这不仅必须用于传统的离线任务的尺寸和设计（例如，确定网络开关点的链路带宽、缓冲容量和处理能力），而且还必须用于在线、性能驱动的流量控制目的，例如会话级别的允许控制。在本提案中，主要研究人员概述了旨在提供分析工具和技术的研究，以分析这些网络及其不同的工作负载，并设计和表征这些网络中的调度策略的属性。主要研究人员建议：1.在网络环境中，在会话可能跨越多个跃点的情况下，开发各种会话性能指标的上限技术，如缓冲区溢出的概率和数据包延迟的分布。2.在处理不同类型（潜在突发）流量源的网络中，开发用于估计缓冲区溢出概率的近似技术；3.开发基于表征最优调度策略的方法的示例路径。

Unlike traditional data networks, future broadband-ISDN (BISDN) wide- area networks will be required to carry a broad range of traffic classes ranging from bursty, variable-rate sources, such as voice and variable- rate coded video, to smooth, constant bit rate sources. Moreover, these networks will have to do so while providing a guaranteed performance or quality-of-service (QOS) to these traffic classes. The problem of characterizing performance in such networks is thus particularly important since this must be done not only for traditional off-line tasks of dimensioning and design (e.g., determining link bandwidths, buffer capacities and processing capacities at network switchpoints) but also for on-line, performance-driven traffic control purposes such as session- level admission control. In this proposal the principal investigators outline research aimed at providing the analytical tools and techniques for analyzing such networks and their diverse workloads and for designing and characterizing the properties of scheduling policies in these networks. The principal investigators propose to: 1.develop techniques to upper bound a variety of per-session performance measures such as probability of buffer overflow and the distribution of packet delay and to do so in a network environment in which sessions may traverse several hops, 2.develop approximation techniques for estimating the probability of buffer overflows in a network handling diverse classes of (potentially bursty) traffic sources, and 3.develop a sample path based on methodology for characterizing optimal scheduling policies.