AF: Small: Collaborative Research: Online Scheduling Algorithms for Networked Systems and Applications

AF：小型：协作研究：网络系统和应用程序的在线调度算法

• 批准号: 0915681
• 负责人: Fei Li
• 金额: --
• 依托单位: George Mason University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2009
• 资助国家: 美国
• 起止时间: 2009-08-01 至 2014-07-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0915681&HistoricalAwards=false
• 关键词: AF; Small; Collaborative; Research; Online

The Internet is now the world's dominant information infrastructure. Numerous requests from Internet users and their applications compete for shared resources in multiple ways. It is therefore critical to efficiently allocate limited network resources in order to provide high quality services. Improving the performance of the Internet in this manner has the potential to have extremely broad impact.Resource management becomes even more challenging when mobile devices connecting to the Internet are considered. Designing efficient algorithms is difficult mainly due to the following factors: (1) diverse and unpredictable resource requests; (2) physical limitations on Internet links, on buffer space in network switches, on capacity of wireless channels, and on battery power in mobile devices.This project aims to provide solutions for several fundamental algorithmic problems in networked systems and applications. Robust and insightful online algorithms will be developed for network switches forwarding prioritized packets and energy management in mobile devices. The objective is to understand the mathematical structure of these problems, to design elegant and easy-to implement online algorithms, to provide rigorous analysis on their performance bounds, and to integrate these algorithms into the real systems to achieve better performance.

互联网现在是世界上占主导地位的信息基础设施。来自互联网用户及其应用程序的大量请求以多种方式竞争共享资源。因此，关键是有效地分配有限的网络资源，以提供高质量的服务。以这种方式提高互联网的性能具有潜在的广泛影响，当考虑到连接到互联网的移动的设备时，资源管理变得更具挑战性。设计有效的算法是困难的，主要是由于以下因素：（1）多样的和不可预测的资源请求;（2）物理上的限制，在互联网链路，在网络交换机的缓冲空间，在无线信道的容量，以及在移动的设备的电池电量。将为移动的设备中的网络交换机转发优先分组和能量管理开发鲁棒的和有洞察力的在线算法。我们的目标是了解这些问题的数学结构，设计优雅和易于实现的在线算法，提供严格的分析其性能界限，并将这些算法集成到真实的系统，以实现更好的性能。