Project COMMUNITY: 2014 Update

Summary of work done in 2012–14

During the first three years of our collaboration, i.e., 2009–2012, we explored new network architectures and proposed mechanisms to enable efficient communication over heterogeneous and disruption-prone networks. Since 2012, we have focused on two main research axes: (1) design of information-centric communication mechanisms for mobile and challenged networks (e.g., delay-tolerant networks), and (2) extension of Software-Defined Networking to facilitate the implementation and large scale deployment of new network architectures, protocols and services in heterogeneous networked environments. We describe our research activities under these two research thrusts below.

New Information-Centric Communication Mechanisms

Recently, a worldwide effort was launched in the network research community to redesign the future Internet’s architecture. Information-Centric Networking (ICN) (also known as Content-Centric Networking or Named Data Networking) is the architecture that attracts most of efforts in the network community. Our contributions related to ICN concern the following items.

  • Extension of CCN to support content dissemination applications in challenging environments. The MeDeHa dissemination mechanism for heterogeneous and challenged networks, we proposed previously was not information-centric. In the context of new ICN architectures, we designed an content-optimal delivery algorithm called CODA for distributing named data over a delay-tolerant network. While disseminating content, CODA maximizes the network throughput by computing the utility of each item published: a device with a full buffer drops content in order of increasing utility and transmits content in order decreasing utility. We integrated CODA within CCN, providing the basic functionalities for querying, caching, and transmitting content [1].
  • Incremental deployment of CCN in the network[3]. We proposed a transition mechanism based on a dynamic overlay optimized for performance and a naming resolution infrastructure that allows Over-The-Top (OTT) services to leverage the use of CCN caches in a global Internet where CCN is only partially deployed. The principle of our solution is to place a gateway speaking CCN and IP at the border of each CCN island and to dynamically build bindings between the name of contents cached in the island and the address of the gateway of the island. The binding is then used to dynamically construct an overlay to interconnect the CCN islands and then leverage the use of caches [3].
  • ICN caching optimization to minimize bandwidth on peering links. We proved that the caching technique proposed in CCN is not optimal in terms of peering links’ bandwidth usage as it results in wastage of limited caching space by duplicating contents throughout the network. Then we proposed a way to deploy an effective caching within an Autonomous System (AS) such that the peering link usage is minimized. We showed with simulations on real topologies that the cost of using off-path caching in terms of delay due to the deflection is compensated by the gain in terms of caching. Indeed, as caching is optimized, a larger number of popular contents can be cached within the AS and hence less traffic exits the AS via the peering links, which results in an overall delay reduction [6].
  • Optimization of CCN caching using SDN. We have evaluated if the SDN approach can help in implementing the optimal caching solution we proposed in [6]. The basic idea is to let the controller decide in an optimal way what are the most popular contents to cache and where, and then to use the OpenFlow controller to update the routing tables of the routers in the network [8]. In this context, we have proposed a solution to improve caching in CCN using an SDN approach that has been presented at the IEEE Infocom student workshop in April 2013 [9].

Software Defined Networking in Heterogeneous Networked Environments

The overarching goal of our activities is to explore the use of SDN techniques, originally proposed for infrastructure-based networks, in future internets that are likely to be even more heterogeneous than today. Our main conributions related to this research thrust are as follows:

  • Secure SDN-Based Capacity Sharing and User Centric Networks (UCNs). We have implemented capacity sharing mechanisms by leveraging SDN in hybrid networked environments, i.e., environments that consist of infrastructure-based as well as infrastructure-less networks. The proposed SDN-based capacity sharing framework provides flexible, efficient, and secure capacity sharing solutions in a variety of hybrid network scenarios. In the paper published at the Capacity Sharing Workshop CSWS 2013 [3], we identify the challenges raised by capacity sharing in hybrid networks, describe our framework in detail and how it addresses these challenges, and discuss implementation issues. To the best of our knowledge, this is the first SDN-based capacity sharing solution that targets hybrid networks and that incorporates security as an integral part of the proposed approach. On recent developments of this work, we published the article [11] on a special issue of Communications Magazine, where we propose the application of our SDN-enabled capacity sharing platform in the context of User Centric Networking (UCN). We discuss user-centric networks as a way of considerably mitigating the problem of sharing limited network capacity and resources efficiently and in a fairly manner. UCNs are self-organizing networks where the end user plays an active role in delivering networking functions such as providing Internet access to other users. We propose to leverage the SDN paradigm to enable cooperation between wireless nodes and provide capacity sharing services in UCNs. Our SDN-based approach allows coverage of existing network infrastructure (e.g., WiFi or 3GPP) to be extended to other end users or ad hoc networks that would otherwise not be able to have access to network connectivity and services. Moreover, the proposed SDN-based architecture also takes into account current network load and conditions, and QoS requirements. Finally, we identify the requirements for enabling capacity sharing services in the context of UCNs: e.g., resource discovery, node admission control, cooperation incentives, QoS, security.
  • Optimizing rule placement in OpenFlow switches. In [12], we advocate that we can trade routing requirements within the network to concentrate on where to forward traffic, not how to do it. As an illustration of the concept, we define an optimization problem that gets the maximum amount of traffic delivered according to policies and the actual dimensioning of the network. The traffic that cannot be accommodated is forwarded to the controller that has the capacity to process it further. We also demonstrate that our approach permits a better utilization of scarce resources in the network.
  • SDN Testbed. We have been built two testbeds, one at Inria and another one at UCSC to evaluate the upcoming SDN-based mechanisms and services. At Inria, the current testbed includes 3 NetFPGA OpenFlow Switches, 3 WiFi Access Points (TP Link supporting OpenFlow), one 48-port Pronto 3290 OpenFlow switch, a Floodlight controller and several mobile laptops. At UCSC, we also have 3 WiFi Access Points (TP Link supporting OpenFlow) as well as 5 linux boxes which will be used as OpenFlow software switches. We also have Android tablets as well as laptops. The two platforms can be connected together to run experimentation scenarios that require both wireless nodes and the Internet infrastructure. We have been working with the Corporation for Education Network Initiatives in California (CENIC) to connect our SDN testbed at UCSC to their SDN testbed, which will connect to the Inria portion of our testbed.
  • Survey of Software Defined Networking (SDN). We have completed a survey on Software Defined Networking [1] which presents early efforts towards programmable networking. The survey also maps the current state-of-the-art in the area as well explores directions of future work. The final document ”was published in early 2014 at IEEE Communications Surveys and Tutorials”.
  • Decentralizing SDN’s Control Plane. We have specified a number of use cases motivating the need for extending the SDN model to heterogeneous networked environments. Such environments consist of infrastructure-based and infrastructure-less networks. These use cases and a high-level description of a Heterogeneous SDN paradigm were summarized in a recent publication [2]. Motivated by the internets of the future that will likely be considerably larger in size as well as highly heterogeneous and decentralized, we sketched out in a short paper [13] a framework aiming to enable not only physical, but also logical distribution of the Software-Defined Networking (SDN) control plane. This framework will accomplish network control distribution by defining a hierarchy of controllers that can “match” an internet’s organizational– and administrative structure. The main idea is to delegate control between main controllers and secondary controllers in order to accommodate administrative decentralization and autonomy. Security should be incorporated as an integral part of the framework.

New Collaborators in COMMUNITY

In the beginning of 2013, Bruno Astuto A. Nunes, former PhD student at UCSC, joined the COMMUNITY associated team as a postdoc. Also, we started a collaboration with Professor Cintia Margi from the University of Sao Paulo (USP) in Brazil. Dr. Margi’s research group brings to the COMMUNITY team an important background on security. They contributed in including the security mechanisms incorporated to our work in [3]. Their work on [10] is another evidence of how they can contribute to our on-going research in terms of developing a secure framework for distributing control, that can be used in challenged scenarios such as in Public Safety Networks. We contend that when distributing the control plane in SDN, security must be integrated to the distributed protocols and is of great importance in challenged environments such as heterogeneous and mobile networks. Two of Dr. Margi’s PhD students, Mateus Santos and Bruno Trevizan, have been working also with us. In June 2013, Mateus started an 1 year internship at UCSC under the supervision of Professor Katia Obraczka.

References

  1. B. Nunes, Marc Mendonca, Xuan-Nam Nguyen, K. Obraczka, Thierry Turletti,”A Survey of Software-Defined Networking: Past, Present, and Future of Programmable Networks“, Communications Surveys & Tutorials, IEEE , vol.16, no.3, pp.1617,1634, Third Quarter 2014, doi: 10.1109/SURV.2014.012214.00180 PDF
  2. B. Nunes, Marc Mendonca, K. Obraczka, Thierry Turletti,”Software Defined Networking for Heterogeneous Networks“, IEEE COMMUNICATIONS SOCIETY MULTIMEDIA COMMUNICATIONS, IEEE COMSOC MMTC E-Letter, Vol. 8, No. 3, pgs 36-39, May 2013. PDF
  3. Mateus A. S. Santos, B. Nunes, Cintia B. Margi, Bruno T. de Oliveira, K. Obraczka, Thierry Turletti,”Software-Defined Networking Based Capacity Sharing in Hybrid Networks“, Workshop on Capacity Sharing (CSWS’13), with the IEEE International Conference on Network Protocols (ICNP13), 2013. PDF
  4. C. Barakat, A. Kalla, D. Saucez, T. Turletti, “Minimizing Bandwidth on Peering Links with Deflection in Named Data Networking“, invited paper to the International Conference on Communications and Information Technology (ICCIT), Beirut, June 2013. PDF
  5. F.M. Ramos Dos Santos, C. Barakat, T. Spyropoulos, T. Turletti, “Content Management in Mobile Wireless Networks“, Research Report http://hal.inria.fr/hal-00742734, 2012. PDF
  6. D. Saucez, A. Kalla, C. Barakat, T. Turletti, “Minimizing Bandwidth on Peering Links with Deflection in Named Data Networking“, Research Report http://hal.inria.fr/hal-00684453, 2012. PDF
  7. D. Saucez, C. Barakat, T. Turletti, “Leveraging Information Centric Networking in Over-The-Top Services“, Research Report http://hal.inria.fr/hal-00684458, 2012. PDF
  8. X.N. Nguyen, “Software Defined Networking in Wireless Mesh Network“, MSc UBINET Thesis, INRIA, August 2012.
  9. X. N. Nguyen; Saucez, D.; Turletti, T., “Efficient caching in content-centric networks using OpenFlow“, INFOCOM Student Workshop, 2013 Proceedings IEEE, pp.1,2, April 2013. PDF
  10. Santos, M. A. S.; Margi, C. B.. “A Secure Multi-Party Protocol for Sharing Valuable Information in Public Safety Networks“. In: Mobile Ad Hoc Networks for Public Safety Systems; Techniques and Challenges in On-Going Work – co-located with IEEE MASS 2011., 2011, Valencia. In Proceedings: 8th IEEE International Conference on Mobile Adhoc and Sensor Systems (IEEE MASS 2011).
  11. B. Nunes, M. A. S. Santos, Bruno T. de Oliveira, K. Obraczka, T. Turletti, C. B. Margi, “Software-defined-networking-enabled capacity sharing in user-centric networks,” Communications Magazine, IEEE , vol.52, no.9, pp.28,36, September 2014, doi: 10.1109/MCOM.2014.6894449 PDF
  12. X.N. Nguyen, D. Saucez, C. Barakat and T. Turletti, “Optimizing rules placement in OpenFlow networks: trading routing for better efficiency”, (long paper), ACM SIGCOMM Workshop on Hot Topics in Software Defined Networking (HotSDN’14), August 2014. PDF
  13. Santos, M. A. S., Nunes, B., Travizan, B., Obraczka, K., Turletti, T., Margi, C. B., “Decentralizing SDN’s Control Plane“, Short paper in: The 39th Annual IEEE Conference on Local Computer Networks (LCN), 2014, Edmonton. PDF

UC Santa Cruz | 1156 High St | Santa Cruz, CA 95064