Sharon Barkai

Knowledge-Defined Networking

The research community has considered in the past the application of Artificial Intelligence (AI) techniques to control and operate networks. A notable example is the Knowledge Plane proposed by D.Clark et al. However, such techniques have not been extensively prototyped or deployed in the field yet. In this paper, we explore the reasons for the lack of adoption and posit that the rise of two recent paradigms: Software-Defined Networking (SDN) and Network Analytics (NA), will facilitate the adoption of AI techniques in the context of network operation and control. We describe a new paradigm that accommodates and exploits SDN, NA and AI, and provide use-cases that illustrate its applicability and benefits. We also present simple experimental results that support, for some relevant use-cases, its feasibility. We refer to this new paradigm as Knowledge-Defined Networking (KDN).

Software defined flow-mapping for scaling virtualized network functions

Network functions such as subscriber-management, content-optimization, security and quality of service, are typically delivered using proprietary hardware appliances embedded into the network as turn-key service-nodes. Next generation functions are being implemented as pure software instances running on standard servers - unbundled components of capacity and functionality. Location-Identity-Separation Overlays enable SDN flow-mapping to dynamically assemble these components to whole solutions by steering the right traffic in the right sequence, of mobile users, to elastic apps & functions.

Global State, Local Decisions: Decentralized NFV for ISPs via Enhanced SDN

The network functions virtualization paradigm is rapidly gaining interest among Internet service providers. However, the transition to this paradigm on ISP networks comes with a unique set of challenges: legacy equipment already in place, heterogeneous traffic from multiple clients, and very large scalability requirements. In this article we thoroughly analyze such challenges and discuss NFV design guidelines that address them efficiently. Particularly, we show that a decentralization of NFV control while maintaining global state improves scalability, offers better per-flow decisions and simplifies the implementation of virtual network functions. Building on top of such principles, we propose a partially decentralized NFV architecture enabled via an enhanced software-defined networking infrastructure. We also perform a qualitative analysis of the architecture to identify advantages and challenges. Finally, we determine the bottleneck component, based on the qualitative analysis, which we implement and benchmark in order to assess the feasibility of the architecture.