|Title||:||Research Topics in Systems and Networking|
|Speaker||:||Aravind Krishnamurthy (University of Washington)|
|Details||:||Tue, 10 Jan, 2017 3:00 PM @ BSB 361|
|Abstract:||:||In this talk, I will describe two research projects that we are
currently pursuing at University of Washington. The first project
focusses on emerging networking architectures that allow for flexible
and reconfigurable packet processing at line rate. These emerging
technologies address a key limitation with software defined networking
solutions such as OpenFlow, which allow for custom handling of flows
only as part of the switchs control plane. Despite their promising
new functionality, flexible switches are not all-powerful; they have
limited state, support limited types of operations, and limit
per-packet computation in order to be able to operate at line rate.
Our work addresses these limitations by providing a set of general
building blocks that mask these limitations using approximation
techniques and thereby enabling the implementation of realistic
The second project focuses on Deep Neural Networks (DNNs), which have become the computational tool of choice for many applications relevant to mobile devices. However, given their high memory and computational demands, running them on mobile devices has required expert optimization or custom hardware. We are developing a framework that, given an arbitrary DNN, compiles it down to a resource-efficient variant at modest loss in accuracy. Further, we introduce novel techniques to specialize DNNs to contexts and to share resources across multiple simultaneously executing DNNs.
About the Speaker:
Arvind is currently Professor of Computer Science and Engineering at University of Washington. He completed his BTech in CSE from IIT-M in 1992 and PhD from Berkeley.