ACM ICN 2020, Montreal, Canada

7th ACM Conference on Information-Centric Networking (ICN 2020)

List of Accepted Papers

  • Far Cry: Will CDNs Hear NDN's Call?

    Chavoosh Ghasemi (University of Arizona), Hamed Yousefi (Aryaka Networks), and Beichuan Zhang (University of Arizona)

    • Abstract:

      During the past two decades, content delivery networks (CDNs) have demonstrated their vital role in providing Internet applications/services with high performance, security, and resiliency. In this paper, we look at CDNs once again, but this time from the eyes of a young technology, called named-data networking (NDN), which promises to solve several old chronic issues of CDNs. Here, for the first time, we scrutinize the NDN standard deployment, i.e., the global NDN testbed, and compare it with two notable commercial CDNs (Akamai and Fastly) at the Internet-scale by deploying an adaptive video streaming service.

      Our experiments reveal the vital role of the forwarding plane in shaping the architecture and determining the performance of these networks. We show that although CDNs’ massive infrastructure provides an excellent quality of experience, NDN’s novel forwarding plane lets the NDN testbed prevail over CDNs in terms of the origin workload and failure resiliency. We also discuss the pros and cons of security models in these networks and show NDN’s potentials for cutting the financial costs of CDNs and enabling unique content retrieval features in the network. Above all, we present several limitations of the current NDN deployment and discuss why the future of NDN hinges on addressing those limitations.


  • PERSIA: a Puzzle-based Interest FloodIng Attack Countermeasure

    Reza Tourani (Saint Louis University), George Torres (New Mexico State university), and Satyajayant Misra (New Mexico State University)

    • Abstract:

      With the proliferation of smart and connected devices, Distributed Denial of Service (DDoS) attacks are increasing. Weak security, improper commissioning, and the fast growth of the IoT industry are the major contributors to the recent DDoS attacks, e.g., Mirai Botnet attack on Dyn and Memcached attack on GitHub. Similar to UDP/TCP flooding (common DDoS attack vector), request flooding attack is the primary DDoS vulnerability in the Named-Data Networking (NDN) architecture. In this paper, we propose PERSIA, a distributed request flooding prevention and mitigation framework for NDN-enabled ISPs. PERSIA’s preventive mechanism eliminates the possibility of successful attacks from malicious end hosts. In the presence of compromised infrastructure (routers), PERSIA dynamically deploys a mitigation strategy to minimize the attack’s magnitude. Our experimentation demonstrates PERSIA’s resiliency and effectiveness in preventing and mitigating DDoS attacks while maintaining legitimate users’ quality of experience (> 99.92% successful packet delivery rate).


  • Liquid Data Networking

    John W. Byers (Boston University) and Michael Luby (ICSI and BitRipple, Inc.)

    • Abstract:

      We introduce Liquid Data Networking (LDN), a design of an information-centric internet whereby all transmitted data is in an encoded, liquid form. Fountain codes are used to efficiently and transparently encode data objects into liquid data, which is generated, transported, dispersed, and cached within the network. The unified LDN framework streamlines transport, load balancing and caching. LDN provides multiple benefits to reliable data object delivery, including reduced delivery latency, increased delivery throughput, faster mobile handoffs, and increased caching capacity at the edge


  • Discovering in-network Caching Policies in NDN Networks from a Measurement Perspective

    Susmit Shannigrahi (Tennessee Technological University), Chengyu Fan (Colorado State University), and Christos Papadopoulos and Craig Partridge (Colorado State University)

    • Abstract:

      Caching is integral to Named Data Networking (NDN). Routers in NDN networks are encouraged to cache content and to serve later requests for content from their caches.

      As NDN has evolved, researchers have come to realize that different caching schemes work better for different types of content and patterns of content requests. From a measurement perspective, this means that being able to determine the caching schemes in use within an NDN network can be essential to understanding the network’s performance.

      In this paper, we investigate the feasibility of detecting NDN caching schemes via active measurement (i.e. by sending requests into the network and measuring responses) from edge systems (e.g. by users). We show it is possible to determine what algorithms routers are using to decide what content to cache. Furthermore, for stochastic caching schemes with fixed caching probabilities, we show it is possible to infer the caching probability. Finally, while we do not seek to understand routers’ cache replacement policies (which we leave to later work), we do find that the methods for determining the caching algorithm are robust to cross traffic that may impact the content of a router’s cache.


  • On the Prefix Granularity Problem in NDN Adaptive Forwarding

    Teng Liang (the University of Arizona), Junxiao Shi (NIST), and Beichuan Zhang (University of Arizona)

    • Abstract:

      One unique architectural benefit of Named Data Networking (NDN) is adaptive forwarding, i.e., the forwarding plane is able to observe data retrieval performance of past Interests and use it to adjust forwarding decisions of future Interests. To be effective, adaptive forwarding assumes what we call Interest Routing Locality, that Interests sharing the same prefix are likely to take the same or similar forwarding path within a short time window, thus past observation can be an indicator of future performance. Since Interests can have multiple common prefixes with different lengths, the real challenge is what prefix length should be used in adaptive forwarding. The longer the common prefix is, the better Interest Path Locality, but the fewer future Interests it covers and the larger the forwarding table size. Existing implementations use static prefix length, which is known to have problems in dealing with partial network failures. In this work we propose to dynamically aggregate and de-aggregate name prefixes in the forwarding table, so to use the prefixes that are most appropriate under the current network situation. To reduce the overhead, we design mechanisms to minimize the use of longest prefix match in the processing of Data packets. Simulations demonstrate that the proposed techniques can make better forwarding decisions under partial network failures with significantly reduced overhead.


  • Leveraging Content Connectivity and Location Awareness for Adaptive Forwarding in NDN-based Mobile Ad Hoc Networks

    Muktadir Chowdhury (University of Memphis), Junaid Ahmed Khan (New York University), and Lan Wang (University of Memphis)

    • Abstract:

      Communication in Mobile Ad-hoc Networks (MANETs) is challenging due to their highly dynamic topology, intermittent connectivity, and low data rate. Named Data Networking (NDN) offers a data-centric approach to communication with an adaptive forwarding plane and in-network data caching, which can be leveraged to address these challenges. In this work, we propose a forwarding strategy called Content Connectivity and Location-Aware Forwarding (CCLF) for NDN-based MANETs. CCLF broadcasts NDN packets and lets each node make independent decisions on whether to forward packets based on per-prefix performance measurements and any available geo-location information. In addition, it employs a density-aware suppression mechanism to reduce unnecessary packet transmissions. Moreover, we have developed a link adaptation layer for ad-hoc links to bridge the gap between CCLF and the capabilities of the underlying link. Our evaluation shows that CCLF not only reduces packet overhead significantly compared to flooding, but also has a data fetching performance close to that achieved by flooding. It also outperforms two other forwarding strategies proposed for information-centric vehicular networks.


  • Named Data Transport: An End-to-End Approach for an Information-Centric IP Internet

    Abdulazaz Albalawi and J.J Garcia-Luna-Aceves (UC Santa Cruz)

    • Abstract:

      Named Data Transport (NDT) is introduced to provide efficient content delivery by name over the existing IP Internet. NDT consists of the integration of three end-to-end architectural components: NDTP (Named Data Transport Protocol), the addition of a new type of resource record in the Domain Name System (DNS) containing manifests describing content, and transparent caches that track pending requests for content. NDT uses receiver-driven Interests to request content and transparent in-network caching of content that enforces privacy, without the need for a clean-slate routing infrastructure. The performance of NDT, the Transmission Control Protocol (TCP), and Named Data Networking (NDN) is compared using off-the-shelf implementations in the ns-3 simulator. The results demonstrate that NDT vastly outperforms TCP and is as efficient as NDN, but without making changes to the existing Internet routing infrastructure.


  • NDN Forwarding at 100 Gbps on Commodity Hardware

    Junxiao Shi, Davide Pesavento, and Lotfi Benmohamed (NIST)

    • Abstract:

      Since the NDN data plane requires name-based lookup of potentially large tables using variable-length hierarchical names as well as per-packet state updates, achieving high-speed NDN forwarding remains a challenge. In order to address this gap, we developed a high-performance NDN router capable of reaching forwarding rates higher than 100 Gbps while running on commodity hardware. In this paper we present our design and discuss its tradeoffs. We achieved this performance through several optimization techniques that include adopting better algorithms and efficient data structures, as well as making use of the parallelism offered by modern multi-core CPUs and multiple hardware queues with user-space drivers for kernel bypass. Our open-source forwarder is the first software implementation of NDN to exceed 100 Gbps throughput, while supporting the full protocol semantics. We also present the results of extensive benchmarking carried out to assess a number of performance dimensions and to diagnose the current bottlenecks in the packet processing pipeline for future scalability enhancements. Finally, we identify future work which includes hardware-assisted ingress traffic dispatching, dynamic load balancing across parallel forwarding threads, and novel caching solutions to accommodate on-disk content stores.


  • iCDN: An NDN-based CDN

    Chavoosh Ghasemi (University of Arizona), Hamed Yousefi (Aryaka Networks), and Beichuan Zhang (University of Arizona)

    • Abstract:

      Despite the close philosophy between content delivery networks (CDN) and named-data networks (NDN), yet no solution has realized a large-scale NDN-based CDN. In this paper, we void the popular belief that any NDN network can be expanded to serve as a CDN and introduce iCDN, a scalable, resilient, and high-performance CDN using NDN technology. iCDN introduces an exclusively customized network design and a novel forwarding strategy for content delivery purposes to let the NDN forwarding plane handle a large body of contents by fully exploiting in-network caches. We evaluate different aspects of iCDN over the Abilene topology and compare it with the global NDN testbed solution to show what makes iCDN a promising solution to build a large-scale NDN-based CDN.


  • What's in a Name? Naming Big Science Data in Information-Centric Networking

    Craig Partridge (Department of Computer Science, Colorado State), Susmit Shannigrahi (Tennessee Technological University), and Chengyu Fan (Colorado State University)

    • Abstract:

      Naming data is the most important construct of Named Data Networking (NDN). The way a piece of content is named has a profound impact on content discovery, routing of user requests, forwarding, retrieval, and security. In addition, and perhaps more importantly, the naming of individual pieces of content seriously affects how the network behaves. While names are ubiquitous in NDN, the design of content names, and how different naming choices affect the network have largely been overlooked. NDN applications and protocols usually name content to fit their particular application scenarios, often derived from existing naming conventions. However, ad-hoc naming schemes often ignore the impact of these names on the network as well as the applications themselves. Our experience in applying NDN to multiple science use cases (e.g., Climate Science, Genomics, and High-energy particle physics) brought forward how different communities name their data, how these names can affect name-based networks and the commonality that exists across these communities. This work points out tradeoffs of different naming schemes, the effect of naming on the network and applications, and finally, provides a set of naming guidelines for future science applications that plan to utilize NDN.


  • Toward a RESTful Information-Centric Web of Things: A Deeper Look at Data Orientation in CoAP

    Cenk Gündogan (HAW Hamburg), Christian Amsüss, Thomas C. Schmidt (HAW Hamburg), and Matthias Wählisch (Freie Universität Berlin)

    • Abstract:

      The information-centric networking (ICN) paradigm offers replication of autonomously verifiable content throughout a network, in which content is bound to names instead of hosts. This has proven beneficial in particular for the constrained IoT. Several approaches, the most prominent of which being Named Data Networking, propose access to named content directly on the network layer. Independently, the IETF CoAP protocol group started to develop mechanisms that support autonomous content processing and in-network storage.

      In this paper, we explore the emerging CoAP building blocks and how they can give rise to an information-centric network architecture for a new RESTful Web of Things. We discuss design options and measure characteristic performances of different network configurations, which deploy CoAP proxies and OSCORE content object security, and compare with NDN. Our findings indicate an almost continuous design space ranging from plain CoAP at the one end to NDN on the other.


  • Connecting the Dots: Selective Fragment Recovery in ICNLoWPAN

    Martine Sophie Lenders (Freie Universität Berlin), Cenk Gündogan and Thomas C. Schmidt (HAW Hamburg), and Matthias Wählisch (Freie Universität Berlin)

    • Abstract:

      In this short paper, we analyze the benefits of integrating 6LoWPAN Selective Fragment Recovery (SFR) in ICNLoWPAN. We present a solution that allows for immediate fragment forwarding—a key feature of SFR—in combination with ICN caching. For this a Virtual Reassembling Endpoint (VREP) is proposed which acts transparently as an SFR fragment forwarder while simultaneously collecting them. Once a datagram is complete, it is exposed to the content cache, effectively making the VREP the new fragmenting endpoint. Our proposal complies with current specs defined in the IETF/IRTF. Furthermore, we offer considerations to combine the reverse path forwarding schemes of both SFR and ICNLoWPAN and assess drawbacks and benefits. In preliminary experiments, we evaluate the network performance of combining ICNLoWPAN with SFR and VREP.


  • ENDN: An Enhanced NDN Architecture with a P4-programmable Data Plane

    Ouassim Karrakchou, Nancy Samaan, and Ahmed Karmouch (University of Ottawa)

    • Abstract:

      Named data networking (NDN) is a content-centric future Internet architecture that uses routable content names instead of IP addresses to achieve location-independent forwarding. Nevertheless, NDN’s design is limited to offering hosted applications a simple content pull mechanism. As a result, increased complexity is needed in developing applications that require more sophisticated content delivery functionalities (e.g., push, publish/subscribe, streaming, generalized forwarding, and dynamic content naming). In this paper, we introduce a novel Enhanced NDN (ENDN) architecture that offers an extensible catalog of content delivery services (e.g., adaptive forwarding, customized monitoring, and in-network caching control) that can be programmed in the data plane using customizable P4 programs. More precisely, the proposed architecture allows hosted applications to associate their content namespaces with a set of services offered by the ENDN control plane. The controller then configures the data plane, which is comprised of two main modules: the enhanced packet processing and the forwarding logic modules. The former parses the packets and queries the enhanced content-based forwarding tables to generate a set of metadata fields used by P4 functions. The latter module is a novel P4 target architecture that executes these P4 functions on the arriving packets. The new architecture extends existing P4 models to overcome their limitations with respect to processing string-based content names. It also allows running independent P4 functions in isolation, thus enabling P4 code run-time pluggability. Experimental results demonstrate the ability of ENDN to achieve network efficiency with low latency.


  • Result Provenance in Named Function Networking

    Claudio Marxer and Christian Tschudin (University of Basel)

    • Abstract:

      Data chunks with names bound to them are "first-class citizens" in information-centric networks. The main service such a network provide to its users is the resolution of names to the associated data. A named function network (NFN) extends this service and also resolves on-demand computation expressions composed from named data and functions. The resolution of computation expressions is completely transparent to the user which makes it very convenient for application developers, however, also means the whole network becomes a “pandora box” that must be trusted totally. In this work we augment NFN with a datastructure that creates transparency about the genesis of every evaluation result. We show that this datastructure enables applications to single out results produced by dubious computing providers and further to maintain trust relationships with these.


  • Analyzing ICN Forwarding Performance on the Wire: A Pragmatic Approach

    Adam Drescher, John DeHart, Jyoti Parwatikar, and Patrick Crowley (Washington University in St. Louis)

    • Abstract:

      Information Centric Networking (ICN) is growing in both popularity and maturity. Two of the main realizations of the concept, Content Centric Networking and Named Data Networking, have both received substantial engineering and academic effort over the years. In particular, an abundance of work focuses on improving the performance of ICN forwarders. Despite this focus, little work has been done on how to evaluate ICN forwarders in a general and rigorous way. This direction is particularly important because ICN forwarders differ substantially from IP forwarders, so much of the pre-existing evaluation literature does not directly apply to ICN. In this paper, we provide a methodology to test the performance of ICN forwarders. Our testing methodology has two key focuses: (i) packet processing performance is the primary metric of exploration, as bytes usually come cheap; and (ii) the performance characteristics of the PIT, FIB, and Content Store are the most important to probe. With these goals in mind, we flesh out experiments that could be run on any similar ICN forwarders. We perform these experiments on NFD and Metis, the two reference forwarders of NDN and CCNx respectively, to reveal conclusions about their performance and behavior under different types of load. Our measurements show Metis can perform just over 15,000 interest/data exchanges per second, and NFD can perform just over 12,000 interest/data exchanges per second. Additionally, we reveal substantial inefficiencies in NFD’s content store due primarily to its ’canBePrefix’ support.