NEST: Efficient Transport of Data Summaries over Named Data Networks

In many emerging data retrieval applications, in response to queries, consumers are interested in getting a summarized version of content quickly rather than retrieving all available data. Recently, Named Data Networks (NDN) have been considered for efficient transfer of summarized information, but the research is still in its infancy. In this paper, we propose NEST, a novel transport protocol for delivering extractive summaries of a dataset distributed across multiple producers over NDN. The goal is to exploit diversity in network conditions between a consumer and different producers towards delivering the consumer-specified summary while minimizing latency. NEST first creates a unified hierarchical representation of the available distributed content using state-of-the-art distributed clustering. Then, using this representation of the dataset, the protocol creates interest messages based on which consumers can opportunistically retrieve representative data objects from the best producers while adapting to dynamic network conditions by capitalizing on the flexibility offered by the NDN infrastructure. We implement NEST on the Mini-NDN network emulator and evaluate its performance using datasets collected from Twitter. Our experimental results show that NEST takes advantage of producer diversity achieving large latency reduction gains of up to 50% compared to baseline protocols.