CAREER: Flow Control and Traffic Modeling: A Rigorous Approach Toward Network Management

职业：流量控制和流量建模：严格的网络管理方法

• 批准号: 9502582
• 负责人: Takis Konstantopoulos
• 金额: $ 13.49万
• 依托单位: University of Texas at Austin
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 1995
• 资助国家: 美国
• 起止时间: 1995-09-01 至 1999-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=9502582&HistoricalAwards=false
• 关键词: CAREER; Flow; Control; Traffic; Modeling

NCR-9502582 Konstantopoulos, Takis This is a proposal about the development of a rigorous approach toward network management and performance analysis for high-speed networks considering local and global aspects of a network management system. Local aspects include the design of flow controllers at the edges of the network that take into account the provision of quality of service and the interaction between users in a network switch. The investigator will extend recently developed techniques that are based on the so-called reflection mapping, develop testing techniques and simulations, and compare these to existing schemes. The investigator will also study traffic modeling. He will develop rigorous methods for validating self-similar traffic processes arising from the operation of an integrated services high-speed networks and use such models in establishing performance analysis methods. Studying the global aspects of a management system involves using a macroscopic picture of the network currently being developed by the investigator. The picture strips out the minutiae of the operation of the network, while retaining its dynamic character (it is not static). The investigator proposes the use of such a picture by a network management system in order to predict, prevent, and correct undesirable behavior that might lead to degradation of services across the network.

NCR-9502582 Konstantopoulos, Takis 这是一项关于为高速网络开发严格的网络管理和性能分析方法的提案，考虑到网络管理系统的本地和全局方面。 本地方面包括网络边缘流量控制器的设计，考虑到服务质量的提供以及网络交换机中用户之间的交互。 研究人员将扩展最近开发的基于所谓反射映射的技术，开发测试技术和模拟，并将其与现有方案进行比较。 研究人员还将研究流量建模。 他将开发严格的方法来验证综合业务高速网络运行所产生的自相似流量过程，并使用此类模型来建立性能分析方法。 研究管理系统的全局方面涉及使用研究者当前正在开发的网络的宏观图景。 该图片去掉了网络运行的细节，同时保留了其动态特征（它不是静态的）。 研究人员建议网络管理系统使用这样的图片来预测、预防和纠正可能导致整个网络服务质量下降的不良行为。