Leveraging Software-Defined Networking for Named Data Networking: Comparison
Please note this is a comparison between Version 2 by Catherine Yang and Version 1 by Rashid Amin.

The internet’s future architecture, known as Named Data Networking (NDN), is a creative way to offer content-based services. NDN is more appropriate for content distribution because of its special characteristics, such as naming conventions for packets and methods for in-network caching. Mobility is one of the main study areas for this innovative internet architecture. The software-defined networking (SDN) method, which is employed to provide mobility management in NDN, is one of the feasible strategies. 

  • software-defined networking
  • OpenFlow
  • Named Data Networking
  • mobility management

1. Introduction

The present network infrastructure is undergoing effective content delivery due to IP-dependent routing, as well as host-to-host exchanges of data in practice and location depending on IP. This innovative tool for the Internet of Things (IoT) addresses these issues. The whole internet architecture has become more sophisticated as a result. Information-centric networking (ICN), a potential future internet design, has been developed to address these challenges. The fundamental principle of the ICN design is to use location-independent naming to separate data (service) from the physical devices storing it [1,2,3][1][2][3]. To improve content distribution for users, ICN offers the chance to switch from IP-based routing to name-based routing, which is independent of location and has a cache for storing material.
ICN seeks to transform the current Internet model from a challenging one to a simple and common one. The distinguished node is certainly not the crucial networking component (servers, switches, terminals). The network conducts all of its operations in accordance with ICN’s defined content objectives. In an ICN design, the router would search for a particular piece of content when a user indicated an interest in it. The user would receive the content if it were discovered. As a result, ICN has a variety of characteristics, including self-secured content, local multicast, name-based routing, and location-independent naming [4].
Content-centric networking (CCN), data-oriented network architecture (DONA), along with network of information (NetInf), as well as the publish-subscribe internet routing paradigm (PSIRP), and NDN are just a few of the ICN architectures that have been presented [5,6,7,8][5][6][7][8]. In order to distribute content efficiently, for example, video streaming, audio streaming, and P2P sharing of data, NDN is considered a future design intended for the internet. It has come to be regarded as an alternative structure for the standard IP-related networking, employing name for routing as a substitute for IP. The data packet contains information like a digital signature, signed documents, and so on. The interest packet also contains information like the name of the seeking material. NDN promotes customer mobility as well. Consumer mobility is experiencing a number of problems, including connection initiation, data packet forwarding, and response delay.
The main goal of ICN is to address IP network restrictions through name-based routing and caching within the network in order to reduce bandwidth use and provide location-independent access to content through several paths as well as mobility management. The fundamental tenet of the present internet architecture is that information is transferred from one point to another and between clients and servers throughout communications. Via specialized routers with caches, which are made up of three tables called the pending interest table (PIT), forwarding information base (FIB), and content store (CS), across which data may be retrieved, NDN shifts the focus from locations (destinations and origins) to the information in itself. As a result, the network should not suffer as of connecting with a server through requesting information in itself [9].
The following are some mobility-related difficulties that the author mentioned in [10]: Mobile devices having mobility features may move between different points of attachment (PoA) with the least amount of handover latency and without affecting the transmission of content, much like in the ICN design. A PoA allows for mobile nodes in order to join a network. Moreover, mobility was therefore split into mobility regarding producers and mobility for consumers. The author claims in [11] that consumer mobility naturally complements the consumer-driven orientation of the NDN. As soon as the mobile user changes to a new PoA, then the I_ packets will be present to perform transmission to keep on typically or to restart after handoff. In contrast to IP designs, the creators of portable content face little difficulties [12].
In order to maintain the availability and location of their content for consumers and center routers, content producers must be able to move their material across locations as quickly as feasible. This is referred to as producer mobility. As a result, the author of [13] concluded that NDN does not aid producer mobility from a unique perspective on mobility assistance. NDN is impacted by a scaling issue with routing table size. Additionally, when providers relocate to a new location, the name system presents major scalability issues [14]. Significant handoff delay and excessive I_packet losses while broadcasting to a producer’s prior location are other problems that need to be rectified [15,16][15][16]. The I_packets continue to follow the prefix trail in the FIB whenever a producer switches locations, thus failing to get to the producer and being dropped. Due to the indirect points utilized to allow producer mobility, the data channel is likewise constrained and difficult. As a result, there is a long handoff interval as a result of conditions.

2. Named Data Networking (NDN) Overview

In order to efficiently distribute material such as streaming videos, music, and P2P services intended for data sharing, NDN is regarded as a future design with respect to the internet. In addition, it uses names to route packets rather than IP addresses and stores data in a cache [6,7,8][6][7][8]. Interest and data packets are the two categories of packets used in NDN. Moreover, the data packet includes information like a digital signature, signed documents, and so on. The interest packet also contains information like the name of the content being sought. The CS, PIT, and FIB are the three tables that make up the NDN node as shown in Figure 21. To store the data locally, CS is utilized, however, the PIT is applied to retain the interface between FIB and the consumer, it is also used as a forwarding strategy as illustrated in Figure 21. Each table has a distinct function. The router verifies the contents of data with the interested party’s name in the CS when they show up at the router for placement of content. The PIT entails an interest list for future data, which is where the router searches if the data is not discovered in the CS. It is sent to the consumer if the operation accomplished on the look-up table has been successful; if not, it is verified with PIT before sending the packet of interest to the source. The router comprising of the whole interests that have conveyed but have not yet been satisfied are kept in the PIT.
Figure 21. Interest vs. data processing approach in NDN [19].
Interest vs. data processing approach in NDN [17].

3. SDN Model for NDN

The division of the network control from the data plane is the fundamental feature of SDN. As illustrated in Figure 32, SDN designs are composed of three layers: the infrastructure layer, the control, and the application layer. On conventional networks, the traffic will be distributed or transmitted, preventing them from having a comprehensive overview of the whole network. On the other hand, because they have logically centralized control, the SDN controller will retain the global perspective of the network.
Figure 32. Software-defined networks model [20].
Software-defined networks model [18].

3.1. A. Infrastructure Layer

The infrastructure layer contains a number of network components that may exchange information and connect with one another, such as switches and routers. These devices carry out two tasks: they gather network status data, temporarily store it, and relay it to the controller. In addition, the second step entails processing the packet in accordance with the precise rules that the controller applies to the packets [20][18].

3.2. B. Control Layer

The management of network traffic falls within the purview of the control layer. It is made up of SDN controllers that allow for logical network control and give an overall picture of the underlying devices of the network. It is the SDN middle or control layer, through which the applications which execute on the controller control the behavior of the infrastructure layer [21][19].

3.3. C. Application Layer

Through various APIs, the application layer offers access points for various services. Northbound interfaces are used on devices to implement numerous APIs that are built for this layer’s diverse functions [21][19].

4. NDN Mobility

The author of the study [5] introduced a brand-new networking model called ICN. Through this research, they look at several crucial aspects, such as techniques for naming as well as routing, along with in-network caching options, and so on, as well as characteristics that distinguish the benefit of actually implementing ICN, unresolved research issues, and fresh appeal in this field. Due to the regularly shifting locations of users, the fundamental help of MLS is in reusing the existing connection regarding a transaction as compared to starting a new connection. The current connection is damaged if a consumer changes their position away from the adjacent node, and the data cannot be entirely received owing to handover. Therefore, in this instance, the connection is reinstated rather than being established anew, and fresh content requests are made to the original new NAR. As a result, the data for the remaining material was delivered by the prior NAR to the NDN producer, who then forwarded it toward the new NAR, which was requested by the user already [17][20]. The transaction is carried out using an IP-based architecture, as designated in the literature [22][21], which generates a simulation atmosphere for the NDN model to analyze performance in case of a mobile consumer changing their position between points to check the rate of generation for the interest packets leveraging and employing ndnSIM within the NS3 simulator. An interest packet generated by a mobile user is initially wrapped with an IP header and sent to the NDN node through the associated AP. The packet is de-encapsulated and routed on the NDN layer when it gets an NDN node, which takes away the IP header from the packet. Additionally, the NDN node delivers IP header-encapsulated data packets to mobile users and returns them [22][21]. The technique will be explained in more depth for the SDN setting when utilizing OpenFlow, as explained in [23][22], where a method is provided that is based on hashing content names with the address field inside IP packets. OpenFlow 1.0 and User Datagram Protocol (UDP) are used to implement this method. The UDP packet’s payload contains the NDN packets that have been encapsulated. Interest and data packets are assigned to two separate port numbers. As a result, the switches may operate with both NDN and IP structures. This creates the routers prepared to detect these two NDN packets with further identification of IP packets from packets in NDN. There is no need in OpenFlow for directing packet payloads, hence NDN routers must be able to conduct longest prefix matching (LPM) on packet names. Consequently, the names are (encrypted), i.e., hashed and added to the packet header’s IP field. The authors in [24][23] demonstrate how the prediction of location can be utilized to perform the anchorless management of mobility and to guarantee a seamless handover for the producer with real-time communication for a multi-media environment. The results show that the methodology reduces round-trip time and handover latency. In [25][24], authors enhanced and optimized a popular scheme for an anchor-based approach via an arrangement of multi-layer anchor nodes based on network topology in a hierarchical way, which is known as an anchor chain. Moreover, the interest packets are forwarded toward the producer, which also pass by it. As the producer changes its location, the newly produced anchor chain can reuse the prior forwarding route to guide further interests according to the modified location for the producer as well, as to reduce the drop rate of the interest packets. In addition, they introduce a mechanism for mobility preprocess that leverage connectionless as well as multi-path packet forwarding within the NDN to create the forwarding path for future interest in advance, that supports the producer to perform a seamless handover. The authors show, using numerical analysis, that their approach reduces latency during the handover and improves response ratio. Similarly, the work in [26][25] suggests a mobility service managed in NDN which is accountable for supervision when a connection is performing a transaction. Hence, the new connection is determined in this manner that is intended for the transaction. Additionally, it hides the connection distortion from the new transaction. Through establishing a mobility service link to the NDN, the mobile service for the consumer is visible to the remaining NDN architecture. Consequently, the consumer’s mobile service, as well as the architecture of NDN can be developed individually. The analytical evaluation and the results from simulation show that the author’s proposed method reduces the transmission of data. Furthermore, the handover latency is compared with respect to the original solution for NDN mobility. In [27][26], the authors proposed a mobility architecture for NDN that leverages SDN. The architecture was illustrated with a single controller and multi-controllers in SDN. However, there were no proof-of-concept experiments to evaluate the validity of the given strategy. Moreover, the procedure was not effectively demonstrated with more details. 

References

  1. Jiménez-Lázaro, M.; Herrera, J.L.; Berrocal, J.; Galán-Jiménez, J. Improving the Energy Efficiency of Software-Defined Networks through the Prediction of Network Configurations. Electronics 2022, 11, 2739.
  2. Ali, J.; Roh, B.-H. An Effective Approach for Controller Placement in Software-Defined Internet-of-Things (SD-IoT). Sensors 2022, 22, 2992.
  3. Siris, V.A.; Ververidis, C.N.; Polyzos, G.C.; Liolis, K.P. Information-Centric Networking (ICN) Architectures for Integration of Satellites into the Future Internet. In Proceedings of the 2012 IEEE First AESS European Conference on Satellite Telecommunications (ESTEL), Rome, Italy, 2–5 October 2012; pp. 1–6.
  4. Vasilakos, A.V.; Li, Z.; Simon, G.; You, W. Information centric network: Research challenges and opportunities. J. Netw. Comput. Appl. 2015, 52, 1–10.
  5. Ahlgren, B.; Dannewitz, C.; Imbrenda, C.; Kutscher, D.; Ohlman, B. A survey of information-centric networking. IEEE Commun. Mag. 2012, 50, 26–36.
  6. Gritter, M.; Cheriton, D.R. An Architecture for Content Routing Support in the Internet. In Proceedings of the 3rd USENIX Symposium on Internet Technologies and Systems (USITS 01), San Francisco, CA, USA, 26–28 March 2001.
  7. Stanford University TRIAD Project, (Online). Available online: www.dsg.stanford.edu/triad/ (accessed on 20 January 2023).
  8. Carzaniga, A.; Wolf, A.L. Content-Based Networking: A New Communication Infrastructure. In Workshop on Infrastruture for Mobile and Wireless Systems; Springer: Berlin/Heidelberg, Germany, 2001; pp. 59–68.
  9. Torres, J.; Ferraz, L.; Duarte, O.C.M.B. Controller-Based Routing Scheme for Named Data Network. Electr. Eng. Program COPPE/UFRJ Tech. Rep. 2012. Available online: https://www.gta.ufrj.br/ftp/gta/TechReports/TFD12.pdf (accessed on 20 January 2023).
  10. Azamuddin, W.M.H.; Aman, A.H.M.; Hassan, R.; Mansor, N. Comparison of Named Data Networking Mobility Methodology in a Merged Cloud Internet of Things and Artificial Intelligence Environment. Sensors 2022, 22, 6668.
  11. Saxena, D.; Raychoudhury, V.; Suri, N.; Becker, C.; Cao, J. Named Data Networking: A survey. Comput. Sci. Rev. 2016, 19, 15–55.
  12. Sinky, H.; Khalfi, B.; Hamdaoui, B.; Rayes, A. Responsive Content-Centric Delivery in Large Urban Communication Networks: A Link NYC Use-Case. IEEE Trans. Wirel. Commun. 2018, 17, 1688–1699.
  13. Zhu, Z.; Afanasyev, A.; Zhang, L. A New Perspective on Mobility Support. NDN Tech. Rep. 2013, 13, 1–6. Available online: http://new.named-data.net/wp-content/uploads/TRmobility.pdf (accessed on 12 January 2022).
  14. Detti, A.; Bracciale, L.; Loreti, P.; Rossi, G.; Blefari Melazzi, N. A cluster-based scalable router for information centric networks. Comput. Netw. 2018, 142, 24–32.
  15. Farahat, H. Proactive Caching to Support Mobility in Named Data Networks. Ph.D. Thesis, Queen’s University, Kingston, ON, Canada, June 2017.
  16. Torres, J.V.; Alvarenga, I.D.; Boutaba, R.; Duarte, O.C.M.B. Evaluating CRoS-NDN: A comparative performance analysis of a controller-based routing scheme for named-data networking. J. Internet Serv. Appl. 2019, 10, 20.
  17. Yi, C.; Afanasyev, A.; Wang, L.; Zhang, B.; Zhang, L. Adaptive forwarding in named data networking. ACM SIGCOMM Comput. Commun. Rev. 2012, 42, 62–67.
  18. Kreutz, D.; Ramos, F.M.; Verissimo, P.E.; Rothenberg, C.E.; Azodolmolky, S.; Uhlig, S. Software-defined networking: A comprehensive survey. Proc. IEEE 2014, 103, 14–76.
  19. Todorova, M.S.; Todorova, S.T. DDoS attack detection in SDN-based VANET architectures. Master’s Thesis, Aalborg University, Aalborg, Denmark, 2016; 175p.
  20. Cha, J.H.; Choi, J.H.; Kim, J.Y.; Han, Y.H.; Min, S.G. A mobility link service for ndn consumer mobility. Wireless Commun. Mobile Comput. 2018, 2018, 5149724.
  21. Liu, Z.; Wu, Y.; Yuepeng, E.; Ge, J.; Li, T. Experimental Evaluation of Consumer Mobility on Named Data Networking. In Proceedings of the 2014 Sixth International Conference on Ubiquitous and Future Networks (ICUFN), Shanghai, China, 8–11 July 2014; pp. 176–472.
  22. Ooka, A.; Ata, S.; Koide, T.; Shimonishi, H.; Murata, M. OpenFlow-Based Content-Centric Networking Architecture and Router Implementation. In Proceedings of the 2013 Future Network & Mobile Summit, Lisboa, Portugal, 3–5 July 2013; pp. 1–10.
  23. Ali, I.; Lim, H. Anchor-Less Producer Mobility Management in Named Data Networking for Real-Time Multimedia. Mob. Inf. Syst. 2019, 2019, 3531567.
  24. Zheng, Y.; Piao, X.; Lei, K. Anchor-Chain: A Seamless Producer Mobility Support Scheme in NDN. In Proceedings of the 2017 IEEE 85th Vehicular Technology Conference (VTC Spring), Sydney, NSW, Australia, 4–7 June 2017; pp. 1–6.
  25. Cha, J.-H.; Choi, J.-H.; Kim, J.-Y.; Min, S.-G.; Han, Y.H. A mobility link service in NDN face to support consumer mobility service. In Proceedings of the 2017 Ninth International Conference on Ubiquitous and Future Networks (ICUFN), Milan, Italy, 4–7 July 2017; pp. 444–449.
  26. Ali, J.; Adnan, M.; Gadekallu, T.R.; Jhaveri, R.H.; Roh, B.H. A QoS-Aware Software Defined Mobility Architecture for Named Data Networking. In Proceedings of the 2022 IEEE Globecom Workshops (GC Wkshps), Rio de Janeiro, Brazil, 4–8 December 2022; pp. 444–449.
More
Video Production Service