blob: d619e360b94e44466bfd035ac484f39b72385e52 [file] [log] [blame]
Alexander Afanasyev4cd50e32013-04-22 18:32:31 -07001ndnSIM research papers
2======================
3
4Here is a list of ndnSIM-related papers. If you want your paper to appear in this list, please send an email to our mailing list or to us directly.
5
6General papers about ndnSIM
7---------------------------
8
9.. note::
10 if you refer to ndnSIM in a published work, please cite this paper, not just the ndnSIM website
11
12- **A. Afanasyev, I. Moiseenko, and L. Zhang, "ndnSIM: NDN simulator for NS-3," NDN, Technical Report NDN-0005, 2012** (`PDF <http://named-data.net/techreport/TR005-ndnsim.pdf>`_, `BibTex <http://lasr.cs.ucla.edu/afanasyev/bibwiki/bibtex/367>`_)
Alexander Afanasyev3fba2882013-07-13 15:43:10 -070013
Alexander Afanasyev4cd50e32013-04-22 18:32:31 -070014 Named Data Networking (NDN) is a newly proposed Internet architecture.
Alexander Afanasyev6ed9f502013-07-19 17:48:35 -070015 NDN retains the Internet's hourglass architecture but evolves the thin waist.
16 Instead of pushing data to specific locations, NDN retrieves data by name.
17 On one hand, this simple change allows NDN networks to use almost all of the Internet's well tested engineering properties to solve not only IP's communication problems but also digital distribution and control problems.
18 On the other hand, a distribution architecture differs in fundamental ways from a point-to-point communication architecture of today's Internet and raises many new research challenges.
19 Simulation can serve as a flexible tool to examine and evaluate various aspects of this new architecture.
20 To provide the research community at large with a common simulation platform, we have developed an open source NS-3 based simulator, ndnSIM, which faithfully implemented the basic components of a NDN network in a modular way.
Alexander Afanasyev4cd50e32013-04-22 18:32:31 -070021 This paper provides an overview of ndnSIM.
Alexander Afanasyev6ed9f502013-07-19 17:48:35 -070022
Alexander Afanasyev4cd50e32013-04-22 18:32:31 -070023
24Research papers that use ndnSIM
25-------------------------------
26
Alexander Afanasyev3fba2882013-07-13 15:43:10 -070027- **L. Wang, A. Afanasyev, R. Kuntz, R. Vuyyuru, R. Wakikawa, and L. Zhang, "Rapid Traffic Information Dissemination Using Named Data," in Proceedings of the 1st ACM workshop on Emerging Name-Oriented Mobile Networking Design - Architecture, Algorithms, and Applications (NoM'12), Hilton Head Island, South Carolina, June 2012, pp. 7–12.** (`PDF <http://lasr.cs.ucla.edu/afanasyev/data/files/Wang/nom.pdf>`_, `BibTex <http://lasr.cs.ucla.edu/afanasyev/bibwiki/bibtex/365>`_, `simulation code <https://github.com/cawka/ndnSIM-nom-rapid-car2car>`_)
Alexander Afanasyev2230d862013-07-04 10:45:26 -070028
Alexander Afanasyev6ed9f502013-07-19 17:48:35 -070029 Our previous work applied the Named Data Networking approach to vehicle-to-vehicle (V2V) communications and developed a simple design for traffic information dissemination applications. This paper uses simulations to evaluate the feasibility of the design as described in [1].
30 Our results show that data names can greatly facilitate the forwarding process for Interest and data packets.
Alexander Afanasyev3fba2882013-07-13 15:43:10 -070031 With adequate vehicle density, data can propagate over long distances robustly at tens of kilometers per second and a requester can retrieve the desired traffic information 10km away in a matter of seconds.
Alexander Afanasyev4cd50e32013-04-22 18:32:31 -070032
Alexander Afanasyev6ed9f502013-07-19 17:48:35 -070033- **Z. Zhu, C. Bian, A. Afanasyev, V. Jacobson, and L. Zhang, "Chronos: Serverless Multi-User Chat Over NDN," NDN, Technical Report NDN-0008, 2012.** (`PDF <http://named-data.net/techreport/TR008-chronos.pdf>`_, `BibTex <http://lasr.cs.ucla.edu/afanasyev/bibwiki/bibtex/371>`_)
Alexander Afanasyev3fba2882013-07-13 15:43:10 -070034
Alexander Afanasyev4cd50e32013-04-22 18:32:31 -070035 Multi-user applications are commonly implemented using a centralized server.
36 This paper presents a new design for multi-user chat applications (Chronos) that works in a distributed, serverless fashion over Named Data Networking.
37 In Chronos, all participants share their views by exchanging the cryptographic digests of the chat room data set.
38 A newly generated message causes a change of the digest at the message originator, which leads to retrieving the new data by all other participants in an efficient way and resynchronization of chat room views.
39 Chronos does not have a single point of failure and eliminates traffic concentration problem of server-based implementations.
40 We use simulations to evaluate and compare Chronos with a traditional server-based chat room implementation.
41 Our results demonstrate Chronos' robustness and efficiency in data dissemination.
42 Chronos' approach of replacing centralized servers by distributed data synchronization can be applied to a variety of distributed applications to simplify design and ease deployment.
43
Alexander Afanasyev3fba2882013-07-13 15:43:10 -070044- **M. Vahlenkamp, "Threats on Information-Centric Networking", Hamburg University of Applied Sciences, Technical Report, 2012.** (`PDF <http://inet.cpt.haw-hamburg.de/teaching/ws-2012-13/master-projekt/markus-vahlenkamp_seminar.pdf>`_)
45
46 The ICN approach aims for reflecting these changes in usage of the Internet and is thus dragging content awareness into the network, for instance to let the network itself decide where to acquire requested data from and thereby utilize content caches to increase the data dissemination efficiency. All this is backed by the use of the publish/subscribe paradigm that is utilised to announce content availability and request its delivery. Since the ICN paradigm is entirely different from todays Internet, new challenges arise within the area of network security. NDN/CCNx, as the most popular ICN approach, claims to solve a couple of different security flaws which the actual Internet is suffering from. This raises the questions of which vulnerabilities still exist and if maybe new issues arise.
47
48- **C. Yi, A. Afanasyev, I. Moiseenko, L. Wang, B. Zhang, and L. Zhang, "A Case for Stateful Forwarding Plane," Computer Communications, vol. 36, no. 7, pp. 779–791, 2013. ISSN 0140-3664** (`PDF <http://lasr.cs.ucla.edu/afanasyev/data/files/Yi/comcom-stateful-forwarding.pdf>`_, `BibTex <http://lasr.cs.ucla.edu/afanasyev/bibwiki/bibtex/380>`_, `simulation code <https://github.com/cawka/ndnSIM-comcom-stateful-fw>`_)
49
Alexander Afanasyev6ed9f502013-07-19 17:48:35 -070050 In Named Data Networking (NDN), packets carry data names instead of source and destination addresses.
Alexander Afanasyev3fba2882013-07-13 15:43:10 -070051 This paradigm shift leads to a new network forwarding plane: data consumers send *Interest* packets to request desired data, routers forward Interest packets and maintain the state of all pending Interests, which is then used to guide *Data* packets back to the consumers.
52 Maintaining the pending Interest state, together with the two-way Interest and Data exchange, enables NDN routers' *forwarding* process to measure performance of different paths, quickly detect failures and retry alternative paths.
53 In this paper we describe an initial design of NDN's forwarding plane and evaluate its data delivery performance under adverse conditions.
Alexander Afanasyev6ed9f502013-07-19 17:48:35 -070054 Our results show that this stateful forwarding plane can successfully circumvent prefix hijackers, avoid failed links, and utilize multiple paths to mitigate congestion.
Alexander Afanasyev3fba2882013-07-13 15:43:10 -070055 We also compare NDN's performance with that of IP-based solutions to highlight the advantages of a stateful forwarding plane.
56
57- **A. Afanasyev, P. Mahadevan, I. Moiseenko, E. Uzun, and L. Zhang, "Interest Flooding Attack and Countermeasures in Named Data Networking," in Proc. of IFIP Networking 2013, May 2013.** (`PDF <http://lasr.cs.ucla.edu/afanasyev/data/files/Afanasyev/ifip-interest-flooding-ndn.pdf>`_, `BibTex <http://lasr.cs.ucla.edu/afanasyev/bibwiki/bibtex/381>`_, `simulation code <https://github.com/cawka/ndnSIM-ddos-interest-flooding>`_)
58
Alexander Afanasyev6ed9f502013-07-19 17:48:35 -070059 Distributed Denial of Service (DDoS) attacks are an ongoing problem in today's Internet, where packets from a large number of compromised hosts thwart the paths to the victim site and/or overload the victim machines.
60 In a newly proposed future Internet architecture, Named Data Networking (NDN), end users request desired data by sending Interest packets, and the network delivers Data packets upon request only, effectively eliminating many existing DDoS attacks.
61 However, an NDN network can be subject to a new type of DDoS attack, namely Interest packet flooding.
Alexander Afanasyev3fba2882013-07-13 15:43:10 -070062 In this paper we investigate effective solutions to mitigate Interest flooding.
63 We show that NDN's inherent properties of storing per packet state on each router and maintaining flow balance (i.e., one Interest packet retrieves at most one Data packet) provides the basis for effective DDoS mitigation algorithms.
64 Our evaluation through simulations shows that the solution can quickly and effectively respond and mitigate Interest flooding.
65
66- **B. Zhou, C. Wu, X. Hong, and M. Jiang, "Algorithms for Distributed Programmable Controllers", Technical Report, March 2013.** (`PDF <http://hong.cs.ua.edu/DCP-techReport-March2013.pdf>`_)
67
Alexander Afanasyev6ed9f502013-07-19 17:48:35 -070068 A few works on SDN (Software-Dened Networking) like those in Onix improve programmability of the distributed network control.
69 The asynchronism and Byzantine issues of the control challenge the re-congurability of the service that is to safely program the control in atomic so as to avoid the transient control issues like the routing loops and black holes.
Alexander Afanasyev3fba2882013-07-13 15:43:10 -070070 We propose two important algorithms of the distributed control to enable the programmability: (1) the reconguration primitive allows the network control of the services being able to safely react to an external event; and (2) the reuse primitive allows the control states of a service being accessible for all services. We give concepts and algorithms of two primitives.
Alexander Afanasyev6ed9f502013-07-19 17:48:35 -070071 In addition, we provide the concrete cases of the current approaches for ICN (Information-Centric Networking) and CDN (Content Distribution Networks) for quests of the recongurability and programmability.
72 Then, we evaluate the performance of ICN in both simulation and the PlanetLab testbed.
73 The evaluation results show that the layer improves the lowers 19.6% of the Interest delays in the ICN that is heavily congested and lowers 97% delays in the PlanetLab with 9 nodes on usual case.
Alexander Afanasyev3fba2882013-07-13 15:43:10 -070074 In addition, the evaluation of CDN on the PlanetLab shows that it reduces 81% request delay on usual case.
75
76- **M. Tortelli, L. A. Grieco, and G. Boggia, "Performance Assessment of Routing Strategies in Named Data Networking", in Proc. of GTTI 2013 Session on Telecommunication Networks, 2013** (`PDF <http://www.gtti.it/GTTI13/papers/Tortelli_et_al_GTTI2013.pdf>`_)
77
Alexander Afanasyev6ed9f502013-07-19 17:48:35 -070078 Information Centric Networking (ICN) architectures are currently being investigated to orient the Future Internet towards a content centric paradigm, thus allowing the provisioning of more secure, efcient, and scalable services.
79 In this work, we focus on the Named Data Networking (NDN) proposal to analyze the impact of several routing and forwarding strategies, which play a fundamental role in ICN.
80 In particular, thanks to the recently devised ns-3 based NDN simulator, namely ndnSIM, we conduce an extensive simulation campaign using the GEANT topology as a core network.
81 We monitor different distinctive metrics, such as le download time, server load reduction, hit ratio, hit distance, and communication overhead, in different topologies and trafc conditions.
82 Simulation results show that the election of a single best forwarding strategy is a difcult task.
Alexander Afanasyev3fba2882013-07-13 15:43:10 -070083 Indeed, the pros and cons of each strategy are heavily inuenced by the popularity distribution of contents, which, in turn, affects the effectiveness of the distributed caching mechanisms typically used in the NDN architecture.
84
Alexander Afanasyev6ed9f502013-07-19 17:48:35 -070085- **S. Seo, J.-M. Kang, A. Leon-Garcia, Y. Han, and J. W.-K. Hong, "Secure and Efficient Context Data Collection using Content-Centric Networking", in Proc. of International Workshop on Smart Communication Protocols and Algorithms (SCPA), 2013** (`PDF <http://dpnm.postech.ac.kr/papers/SCPA/13/sesise/scpa13.pdf>`_)
Alexander Afanasyev3fba2882013-07-13 15:43:10 -070086
Alexander Afanasyev6ed9f502013-07-19 17:48:35 -070087 Context data collection is a fundamental and important process for realizing context-aware recommender or personalization systems.
88 The existing context data collection approaches are based-on traditional TCP/IP that has several disadvantages such as lack of mobility and security.
89 On the other hand, Content-Centric Networking (CCN) provides advantages in terms of mobility, security, and bandwidth efciency compared to TCP/IP.
90 In this paper, we propose a secure and efcient context data collection and provision approach based on CCN.
Alexander Afanasyev3fba2882013-07-13 15:43:10 -070091 Simulation results show that this approach can reduce bandwidth consumption by 52.7%–98.9% in comparison to a TCP/IP-based one.
Alexander Afanasyev6ed9f502013-07-19 17:48:35 -070092
93- **J. Ran, N. Lv, D. Zhang, Y. Ma, and Z. Xie, "On Performance of Cache Policies in Named Data Networking", in International Conference on Advanced Computer Science and Electronics Information (ICACSEI 2013), 2013** (`PDF <http://www.atlantis-press.com/php/download_paper.php?id=7640>`_)
94
95 Named Data Network (NDN) is gaining increasingly concerns, as an important direction of the future Internet architecture research centered on content.
96 Content caching has played a key role in NDN.
97 Existing cache replacement policies like Least Frequently Used (LFU) and Least Recently Used (LRU) have failed to make full use of the popularity of contents, which leads to a low cache efficiency in the dynamic network.
98 In order to make the best use of content popularity in the cache strategy, this paper has proposed a cache replacement policy based on content popularity (CCP), and designed the data structure format and replacement algorithm.
99 For fully studying and analyzing the performance of different cache policies in NDN in terms of network throughput, server load and cache hit ratio, we have done a lot of simulations to show how they will improve the network.
100 The simulation results show that our proposed CCP can significantly decrease the server load with a higher cache hit ratio and increase the network capacity at the same time compared with LRU and LFU.
101 And the average throughput is reduced significantly by nearly 47% in comparison to that of the case without in-networking caching.
102 Moreover, it also shows the performance under different sizes of content store.
103 The effectiveness of the CCP strategy is proved during the simulation.
Alexander Afanasyeve63a7c12013-07-30 11:25:50 -0700104
Alexander Afanasyev772f51b2013-08-01 18:53:25 -0700105- **M. Wahlisch, T.C. Schmidt, and M. Vahlenkamp, "Backscatter from the Data Plane--Threats to Stability and Security in Information-Centric Network Infrastructure", in Computer Networks, 2013** (`DOI 10.1016/j.comnet.2013.07.009 <http://dx.doi.org/10.1016/j.comnet.2013.07.009>`_)
Alexander Afanasyeve63a7c12013-07-30 11:25:50 -0700106
107 Information-centric networking (ICN) raises data objects to first class routable entities in the network and changes the Internet paradigm from host-centric connectivity to data-oriented delivery.
108 However, current approaches to content routing heavily rely on data-driven protocol events and thereby introduce a strong coupling of the control to the data plane in the underlying routing infrastructure.
109 In this paper, threats to the stability and security of the content distribution system are analyzed in theory, simulations, and practical experiments.
110 We derive relations between state resources and the performance of routers, and demonstrate how this coupling can be misused in practice.
111 We further show how state-based forwarding tends to degrade by decorrelating resources.
112 We identify intrinsic attack vectors present in current content-centric routing, as well as possibilities and limitations to mitigate them.
113 Our overall findings suggest that major architectural refinements are required prior to global ICN deployment in the real world.