|
Nick Feamster Assistant Professor Networking Group School of Computer Science College of Computing Georgia Tech Klaus Advanced Computing Building Room 3348 feamster - gatech . edu + 1 404 385 1944 I do not check voice mail. CV (April 2008) Publications Bio |
Improving Internet Availability
An early paper on path splicing
appeared at HotNets 2007; a full version will appear at ACM SIGCOMM
2008.
Network Security
Our work on improving Internet accountability appeared at HotNets 2007;
a full version will appear at ACM SIGCOMM 2008.
Our work on behavioral blacklisting for
spam filtering appeared at ACM CCS 2007.
Our paper on spam received
the Best Student Paper award at SIGCOMM 2006.
Network Troubleshooting
Our paper on evaluating "What-If" scenarios in network configurations
will appear at ACM SIGCOMM 2008.
Our paper on network-wide disruption
detection appeared at SIGMETRICS 2007.
My research develops tools, techniques, and protocols to improve the availability and performance of communications networks in the face of failures, misconfiguration, and malice. Simply put, I try to help network operators do their jobs better.
Research area. My research focuses on networked computer systems, with a strong emphasis on (1) network operations; (2) network architecture and protocol design; (3) high performance (i.e., high availability, high throughput) wired and wireless networks; and (4) anti-censorship techniques and systems.
Goal. The primary goal of my research is to help network operators run their networks better, and to enable users of these networks to experience high availability and good end-to-end performance. I am strongly interested in tackling practical, real-world problems using a ``first principles'' approach, designing systems based on these principles, and implementing and deploying these systems in practice.
Approach. My research runs "from practice, to theory, back to practice". I look to the real world for inspiration and practical problems. I then design solutions to these problems that have provable properties and solid theoretical backing. Finally, I build and deploy real systems based on these solutions. This first principles approach means that I bring many "tools" to bear, from algorithms to economics to machine learning. I place a strong emphasis on transfer of these results back to practice: the resulting tools and algorithms have been adopted or applied in practice.
More details.
Please see this research statement
(Feb. 2007) for an overview of the research areas where my students
and I are most active.
Or, see my Three-Minute Madness Slides (August 17, 2007)
I also serve on the North American Network Operators Group (NANOG) Program Committee. Part of my role is to bring interesting networking research to the network operations community (i.e., operators and vendors). If you think that sounds like you, please feel free to contact me.
Masters Students
Bilal Anwer
Kaushik Bhandankar
Chris Kelly - campus network
operations
Maria Konte
Yogesh Mundada