Preliminary Program

The workshop will feature the following talks: (Room Number: Willow Glen III
  • Opening Remarks: "The Goals for the Workshop: Edge of Networking Computing in Wired and Wireless Networks" by Shahrokh Daijavad, Sumedh Sathaye and Jane Xu, IBM T.J. Watson Research Center "
  • "OpenFlow/Software-Defined Networking: Enabling Network Innovations" by Guru Parulkar, Executive Director Clean Slate Internet Design Program, and, Consulting Professor of Electrical Engineering, Stanford University

  • OpenFlow/Software Defined Networking (SDN) is a new approach to networking that has the potential to enable on-going network innovation in a production setting. Key aspects of SDN include: separation of data and control planes; a uniform vendor agnostic interface called OpenFlow between control and data planes; logically centralized control plane, realized using a network OS, that constructs and presents a logical map of the entire network to services or control applications on top; and slicing and virtualization of the underlying network. With SDN a researcher, network administrator, or third party can introduce a new capability by writing a software program that simply manipulates the logical map of a slice of the network. Researchers around the world are starting to deploy SDN networks for research and limited production use. SDN also forms the network substrate of NSF's GENI infrastructure designed to enable research at scale in networking and distributed systems. In this talk I will share the SDN story (so far): rationale, design, deployments, and coming together of an ecosystem.
  • "Cloud and Fog Computing: Trade-Offs and Applications" by Flavio Bonomi, Cisco Fellow and Vice President, Advanced Architecture and Research Organization at Cisco Systems

Over the past years, two major technology trends have gained immense relevance, and have shaped human experience. The first is the development of smart handheld devices, rich in computing, storage and networking capabilities. The second is the evolution of large Data Centers and Cloud Services, which are delivering a vast array of services to the end users.
We believe there is a missing link towards the fulfillment of the potential for a more scalable, efficient, and performing
end-to-end IT platform, which satisfies the growing desires of end users, but also will enable more broadly the connection and control of complex systems such as buildings, cities, health care and transportation.
The missing link is what we started calling “Fog Computing”, a platform providing more distributed availability of computing and storage resources at the edge of the network, closer to the end-points.
In this talk, we will explore some of the trade-offs between cloud and fog computing, and describe some applications benefiting from this new architecture.

  • "Adaptive Remote Execution Scheduler for Multithreaded Mobile Devices" by  Tomer M. London, Asaf Cidon, Sachin Katti, Christos Kozyrakis, Mendel Rosenblum. Stanford University
Remote execution systems enable a mobile device to offload computations to a remote server or cloud, in order to enhance its computing capabilities or reduce its energy consumption. Such systems utilize a scheduler to dynamically decide where to execute each remote execution RPC, while adapting to changing network conditions and application demands. We present an adaptive remote execution scheduler for multithreaded mobile devices that enables mobile device operating systems to seamlessly schedule tasks on a remote server as if it were a local resource, where the wireless network serves as the interconnect.

  • "Building Network Edge Intelligence for Virtualized Data Centers" by Vijoy Pandey, CTO of Blade Network Technologies

  • "Wireless and IT Convergence at the Edge of the Network + Internet of Things" by Shahrokh Daijavad, Sumedh Sathaye and Jane Xu, IBM T.J. Watson Research Center
The tremendous growth in consumer traffic in the wireless networks and the emerging "smarter planet" traffic from smarter meters, smarter grid, smarter homes, etc. has necessitated adding "intelligence" to the edge of the network by bringing IT applications and "Analytics" closer to the mobile users. These applications can either perform real-time, very low-latency functions on "data in motion" or can be paired with their counterpart applications in the core network data centers, operating on massive amounts of "data at rest". Consequently, traditional wireless base band computations performed at such network edges as Base Stations will be running side-by-side with IT applications on the same physical platform. This convergence of IT and wireless computations creates interesting new requirements on processors and systems that are used at the edge.
  •  Panel Discussion: "Edge of the Network Computing, A New Battleground for IT, Network, Wireless/NEP Companies?"