While the Internet has far exceeded expectations, it has also stretched initial assumptions, often creating tussles that challenge its underlying communication model. Users and applications operate in terms of content, making it increasingly limiting and difficult to conform to IP's requirement to communicate by discovering and specifying location. To carry the Internet into the future, a conceptually simple yet transformational architectural shift is required, from today's focus on where ? addresses and hosts ? to what ? the content that users and applications care about. This project investigates a potential new Internet architecture called Named Data Networking (NDN). NDN capitalizes on strengths ? and addresses weaknesses ? of the Internet's current host-based, point-to-point communication architecture in order to naturally accommodate emerging patterns of communication. By naming data instead of their location, NDN transforms data into a first-class entity. The current Internet secures the data container. NDN secures the contents, a design choice that decouples trust in data from trust in hosts, enabling several radically scalable communication mechanisms such as automatic caching to optimize bandwidth. The project studies the technical challenges that must be addressed to validate NDN as a future Internet architecture: routing scalability, fast forwarding, trust models, network security, content protection and privacy, and fundamental communication theory. The project uses end-to-end testbed deployments, simulation, and theoretical analysis to evaluate the proposed architecture, and is developing specifications and prototype implementations of NDN protocols and applications.
As one of the four projects under NSF's Future Internet Architecture Program, the Named Data Networking (NDN) project set out to address many of the major shortcomings of the current Internet by starting with a fundamentally different premise. Rather than the host-based, point-to-point communication model employed today, NDN names data instead of locations, and organizes its architectural decisions around models of communication driven primarily by data distribution and the interests of end users. The NDN team consisted of a diverse mix of over 20 researchers from 10 campuses bringing a wide spectrum of expertise to tackle the ambitious interdisciplinary reserarch agenda. CAIDA researchers contributed to activities in two project areas: Routing and Forwarding, and Evaluation and Measurement. Starting in 2013, CAIDA personnel also provided overall management support to the project. In Routing and Forwarding, our primary research contribution focused on exploring the most ambitious routing research idea that emerged from NSF's FIND program: greedy routing based on underlying metric spaces. We developed HyperMap, a simple method to map a given real network to its hyperbolic space by replaying the network's geometric growth. The implemented algorithm estimates the hyperbolic coordinates of new nodes at each time step by maximizing the likelihood of the network snapshot in the model. HyperMap outperforms our previous embedding methods in terms of mapping accuracy, method simplicity, and computational complexitiy. We applied HyperMap to embed the AS-level Internet topology derived from CAIDA's Ark measurements into its hyperbolic space and thus obtained hyperbolic coordinates of the ASes participating in the NDN testbed. We conducted routing experiments on the NDN Testbed investigating the performance metrics for the modified greedy forwarding (MGF) algorithm that excludes the current node from any distance comparisons and finds the neighbor closest to the destination. We measured the success ratio (the percentage of the successful paths that reach their destinations) and the average stretch (the ratio of the hop lengths of greedy paths to the corresponding shortest paths in the graph). We simulated forwarding on the full graph of participating sites as well as on all the graphs obtained from the full graph by removing one link without disconnecting the full graph. Our experiments demonstrated high efficiency and robustness of greedy forwarding when using the underlying hyperbolic metric space to calculate the distance between participating nodes. We developed hggraphs, a C++ library that provides a collection of functions and data structures for generating synthetic graphs embedded in hyperbolic metric spaces, and computing properties of those graphs. This library supports research and development of hyperbolic routing in the NDN environment as it enables the implementation of tools to assess the effectiveness of the greedy routing approach in synthetic networks of variable size. It also allows the researchers to create new ndnSIM scenarios extending the default forwarding strategy to simulate hyperbolic routing. Two postdocs participated in NDN project activities in Routing and Forwarding area and developed software for routing research and simulations. As part of Testing and Evaluation activities, we maintained a local node on the national NDN testbed using the CCNX hub software. We also parrticipated in team experiments testing NDN-based video and audio software, participatory sensing, and media distribution via the NDN infrastructure. Finally, during the last two years of the project, we provided consistent and efficient management support for the whole NDN team, overseeing and coordinating the activities of all participating institutions. CAIDA personnel organized and conducted weekly management conference calls and monthly calls for all area PIs, tracked action items, and sent minutes to the NDN PI and/or NDN participants mailing lists. We hosted and maintained the NDN project wiki, edited and posted online annual project reports, drafted and posted an NDN FAQ, and contributed to the design of the new NDN web site launched in 2013. We participated in strategy meetings, co-organized and hosted two project retreats, and co-organized the first NDN community meeting.
