This project proposes a comprehensive load-balancing solution to minimize client response time and reduce system costs for services hosted in wide-area networks. The system, called Aster*x, uses the global state of server load and network congestion, and dynamically routes the requests over appropriate (server and path) pairs, calculated using the load-balancing algorithms developed by project staff.
The GENI network infrastructure will be used for extensive deployment, evaluation, and demonstration of Aster*x. Aster*x exploits OpenFlow?s logically centralized controller to obtain the global network state and route flows of various granularities. It will use the PlanetLab and ProtoGENI-based computation substrate to host the replicated web service and to generate client requests from multiple locations. The project will provide an opportunity for students across four universities to collaborate and build a relatively large experimental system on GENI.
Most of the work on this project was for the support of the Stanford Aster*x project. That is primarily being provided by PIs Feamster and Clark. Because the funding far outlasted the originally proposed work, we were able to use the research funds to achieve significantly more than originally planned. In particular, we were able to develop tools and techniques to help improve the user experience and application performance in home networks, through several additional projects: The development of an initial prototype of a Software Defined Networking-based system that helps users to monitor and manage data usage caps in their home networks. The EAGER project funded some of the initial work on this effort, including a Ph.D. student, which subsequently became a full research project that is currently in submission to ACM CHI. The development of an algorithm and system (and accompanying software release) that improves the performance of web browsing for home users by as much as 30%. This work resulted in a publication in ACM SIGCOMM Internet Measurement Conference and a major software release. The development an algorithm and tool to help users debug the performance of their home wireless networks. This work resulted in a major software release and deployment (in cooperation with the Federal Communications Commission) to more than 4,000 home networks, as well as an initial publication on the work in Passive and Active Measurement. The intellectual merit of the work involves the development of new algorithms for characterizing and improving the performance of both home wireless networks and applications in home broadband access networks (e.g., Web, streaming video). The broader impact of the work involves the development of publicly available software tools---including a deployment of several of these tools in more than 4,000 home networks across the United States---in collaboration with the Federal Communications Commission. As consumers increasingly rely on the Internet for critical services, understanding and improving the performance of last-mile access networks is paramount. The resaerch that this EAGER grant initiated continues under CNS-1422680.
