Networking for Big Data: Theory and Optimization for NDN

The advent of Big Data is stimulating the development of new networking architectures which facilitate the acquisition, transmission, storage, and computation of data. In particular, Named Data Networking (NDN) is an emerging content-centric networking architecture which focuses on enabling end users to obtain the data they want, rather than to communicate with specific nodes. By naming content instead of their locations, NDN transforms data into a first-class network entity.

In this talk, we present a new analytical and design framework for the optimization of key network functionalities within the NDN architecture, which is also broadly applicable to content delivery and peer-to-peer networks. The framework includes the joint optimization of traffic engineering and caching strategies, in order to best utilize both bandwidth and storage for efficient content distribution. It also includes optimal congestion control when user demand for content becomes excessive. We first develop distributed and adaptive algorithms for joint request forwarding and dynamic cache placement and eviction, which effectively achieve network load balancing, thereby maximizing the user demand rate that the NDN network can satisfy. Next, we develop content-based congestion control algorithms which naturally work in concert with forwarding and caching to achieve a favorable tradeoff between the aggregate user utility from admitted content requests and the total user delay. Numerical experiments within a number of network settings demonstrate the superior performance of these algorithms in terms of multiple metrics.

Joint work with Tracey Ho, Ying Cui, Ran Liu, Michael Burd, and Derek Leong