[PAST EVENT] Building Flexible, Dynamic Networks with Software Defined Network Function Virtualization

September 26, 2014
3pm
Location
McGlothlin-Street Hall, Room 020
251 Jamestown Rd
Williamsburg, VA 23185Map this location
Building Flexible, Dynamic Networks with Software Defined Network Function Virtualization

Timothy Wood, George Washington University

Recent improvements in multi-core processors and high-speed network interface cards have enabled Network Function Virtualization (NFV), which allows traditional network hardware such as routers, firewalls, and switches to instead run on commodity servers. NFV makes the network data processing elements run as software, allowing them to be deployed dynamically and easily tuned with changes in network workloads. At the same time, Software Defined Networking (SDN) has grown in popularity as a way to more easily manage network services by centralizing control plane functions.

Our work is exploring how the intersection of these two technologies can enable a new breed of highly dynamic network services in both wide area and data center networks. We have developed the NetVM platform, which allows network software running in virtual machines to support speeds of 10 Gbps and beyond through zero-copy memory mechanisms and careful NUMA-aware processor management. This talk will discuss the challenges at the systems and resource management levels needed to achieve the vision of a flexible network infrastructure where service components can be dynamically instantiated on demand, and flows can be intelligently rerouted based on complex application and flow state that is not currently visible to network controllers.

Biography: Timothy Wood is an assistant professor in the Department of Computer Science at George Washington University. Before joining GW he received a doctoral degree in computer science from the University of Massachusetts Amherst and a bachelor's degree in electrical and computer engineering from Rutgers University. His research studies how cloud computing platforms can be built from massive data centers containing thousands of servers, storage, and network devices. He seeks to improve the performance, reliability, and energy efficiency of these large distributed systems by adding automation and intelligence at the operating system and virtualization layers. His PhD thesis received the UMass CS Outstanding Dissertation Award, his students have voted him CS Professor of the Year, and he has won two best paper awards, a Google Faculty Research Award, and an NSF Career award.