Short Talks

Planning Hybrid SDN and Legacy Networks with Constraint Satisfaction

Sanjai NarainDr. Sanjai Narain

Fellow and Chief Scientist, Information Assurance and Security Department

Applied Communication Sciences (ACS)


Despite the great potential of software-defined networking, its assimilation into legacy networks is likely to be gradual. Network operators will likely replace parts of their networks with SDN to gain experience with it and understand how its strengths can be combined with those of legacy networks. Thus, tools are needed to conceptualize overall security and functionality requirements of a network and plan how these can be satisfied using an SDN part and a legacy part as appropriate. This talk presents a suite of such tools called Distributed Assured and Dynamic Configuration. These implement a Science of Configuration using SMT solvers.

A Low-Latency Network Monitoring Platform

UnknownJeff Rasley

Ph.D. student 

Brown University 


Software-defined networks allow us to build self-tuning networks that provide the ability to monitor and react to network conditions such as congestion. Unfortunately, state-of-the-art network monitoring methods require 100s of milliseconds to seconds to extract global network state, such as link utilization and flow throughput information.
This talk discusses Planck, a network measurement architecture that employs oversubscribed port mirroring to extract network information 1-2 orders of magnitude faster than recent approaches. I will show the utility of this mechanism in the contexts of traffic engineering, tenant-level performance isolation, and network debugging.

Simon: Scriptable Interactive Monitoring for SDNs



Da Yu

Ph.D. Student

Brown University



Although Software-Defined Networking can simplify network management, it also poses new testing and debugging challenges for operators. Debugging is often an interactive process that involves stepping through data and control-plane events and performing actions in response. Sometimes, however, this interactive process can become highly repetitive; in such cases, we should be able to script the activity to reduce operator overhead and increase reusability.

We introduce SIMON, a Scriptable Interactive Monitoring system for SDN. With SIMON, operators can probe their network behavior by executing scripts for debugging, monitoring, and more. SIMON is independent of the controller platform used, and does not require annotations or intimate knowledge of the controller software being run. Operators may compose debugging scripts both offline and interactively at SIMON’s debugging prompt. In the process, they can take advantage of the rich set of reactive functions SIMON provides as well as the full power of Scala. We present the design of SIMON and discuss its implementation and use.