FOGROUTE: DTN-BASED DATA DISSEMINATION MODEIN FOG COMPUTING
ABSTRACT
Fog computing, known as “cloud closed to ground”,deploys light-weight compute facility, called Fog servers, atthe proximity of mobile users. By pre-catching contents in theFog servers, an important application of Fog computing is toprovide high-quality low-cost data distributions to proximitymobile users, e.g., video/live streaming and ads dissemination,using the single-hop low-latency wireless links. A Fog computingsystem is of a three tier Mobile-Fog-Cloud structure; mobile usergets service from Fog servers using local wireless connections,and Fog servers update their contents from Cloud using thecellular or wired networks. This, however, may incur high contentupdate cost when the bandwidth between the Fog and Cloudservers is expensive, e.g., using the cellular network, andistherefore inefficient for non-urgent, high volume contents. Howto economically utilize the Fog-Cloud bandwidth with guaranteeddownload performance of users thus represents a fundamentalissue in the Fog computing. In this paper, we address the issueby proposing a hybrid data dissemination framework whichapplies SDN (software defined network) and DTN (delay tolerablenetwork) approaches in Fog computing. In specific, we decomposethe Fog computing network with two planes, where the cloud is acontrol plane to process content update queries and organize dataflows, and the geometrically distributed Fog servers form a dataplane to disseminate data a Fog servers with delay tolerantnetwork technique. Using extensive simulations, we show thatproposed framework is efficient in term of data disseminationsuccess ratio and content convergence time a Fog servers.
EXISTING SYSTEM:
Fog computing is an emerging paradigm toextend the Cloud service to the “ground”, where applicationsor contents located on far away worldwide Cloud servers arenow distributed to a multitude of Fog servers located just inlocal business premiss, on-board or even the outdoor, whichcan also be used in urban computing. These Fog servershave process, storage and network transmission capabilities.Mobile users access Fog servers with just one-hop wirelessconnection, and therefore streaming applications for majorityof mobile users become available even at the area in a poorInternet coverage. The transmission speed from Fog server tomobile user is much faster than the speed from remote Cloudto end users.For high volume and non-urgent data, it is not wise touse expensive network transmission techniques. Instead, delaytolerant network technique could be used to provide datadissemination between Fog servers, and between mobile usersand Fog servers as well.DelayTolerant Network (DTN) is featured withlong latency and unstable network topology, where end-to-enddelay can be measured in hours or days and data disseminationpaths may not exist. Data dissemination in DTN uses storecarry-forwardtechniques; if there is no connection availableat a particular time, the source node needs to store and carrythe message until the next available node is encountered.
PROPOSED SYSTEM:
In summary, thecontributions of this paper are three fold:•Practical system: we have developed a practical Fogcomputing based content dissemination framework invehicular networks. The proposal makes use the delaytolerant model and also the Cloud computing framework.•Distributed algorithm: we have developed a fully distributedalgorithm to enable the low-latency and low-costcontent disseminations in the vehicular fog computingsystem.• Performance evaluation: we have conducted extensivesimulations to verify the performance of the proposal.It is show that our proposal can achieve a low endto-enddelay of data dissemination with high successfulprobability.
CONCLUSION AND FUTURE WORK
In this paper, we have proposed a hybrid data disseminationmodel in Fog computing by utilizing both delay tolerantnetwork and Cloud techniques. These contents with highervolume and lower timeliness, such as high definition video,could be disseminated by delay tolerant technique. While thesecontents with lower volume and higher timeliness, such asemergency news, could be disseminated with current Cloudtechnique. The proposed model is central controlled by contentprovider or Cloud administer and all Fog servers aredistributed deployed in local areas. Delay tolerant networkbased data dissemination is based on human (mobile devices)and vehicles. Experiment results demonstrated the proposedmodel has a higher delivery success ratio and lower end to enddelay, which proved it is not only workable, but also reliableand efficient in term of data dissemination in Fog computing.For the future work, we intend to extend the current proposalto a more generalized scenario as follows. The currently carrierforwarding algorithm in delay tolerant network is based onone hop delivery only, where carriers download content fromcontent provider (source Fog server), carry to the destinationFog server and upload the content to it. In that case, if amobile device has limited connect with the destination Fogserver, it cannot be selected as carrier. In the future work, theforwarding algorithm could be designed as relay-exchangeforwarding,where mobile devices could exchange/multi-castcontents a themselves when they are moving to the finaldestination.
REFERENCES
[1] F. Bonomi, R. Milito, J. Zhu, and S. Addepalli, “Fog computing andits role in the internet of things,” in Proceedings of the First Edition ofthe MCC Workshop on Mobile Cloud Computing, ser. MCC ’12. NewYork, NY, USA: ACM, 2012, pp. 13–16.
[2] Cisco, “Cisco delivers vision of fog computing to accelerate value frombillions of connected devices,” Cisco Press Release, Jan 2014.
[3] T. H. Luan, L. Gao, Z. Li, Y. Xiang, and L. Sun, “Fog Computing:Focusing on Mobile Users at the Edge,” ArXiv e-prints, Feb. 2015.
[4] C. W. Paper, “Cisco visual networking index: Global mobile data trafficforecast update, 2014?019,” Feb 2015.
[5] I. Stojmenovic and S. Wen, “The fog computing paradigm: Scenariosand security issues,” in Computer Science and Information Systems(FedCSIS), 2014 Federated Conference on, Sept 2014, pp. 1–8.
[6] R. Deng, R. Lu, C. Lai, T. H. Luan, and H. Liang, “Optimal workloadallocation in fog-cloud computing towards balanced delay and powerconsumption,” IEEE Internet of Things Journal, May 2016.
[7] Z. Xu, W. Liang, W. Xu, M. Jia, and S. Guo, “Efficient algorithms forcapacitated cloudlet placements,” IEEE Transactions on Parallel andDistributed Systems, vol. 27, no. 10, pp. 2866 – 2880, October 2015.
[8] Y. Zheng, L. Capra, O. Wolfson, and H. Yang, “Urban computing:Concepts, methodologies, and applications,” ACM Trans. Intell. Syst.Technol., vol. 5, no. 3, pp. 38:1–38:55, Sep. 2014.
[9] L. Gao, S. Yu, T. H. Luan, and W. Zhou, Delay Tolerant Networks andTheir Applications. Springer Publishing Company, Incorporated, 2015.
[10] Y. Cao and Z. Sun, “Routing in delay/disruption tolerant networks: Ataxonomy, survey and challenges,” Communications Surveys Tutorials,IEEE, vol. 15, no. 2, pp. 654–677, Second 2013.
[11] A. Pentland, R. Fletcher, and A. Hasson, “DakNet: rethinking connectivityin developing nations,” Computer, vol. 37, no. 1, pp. 78–83, 2004.
[12] L. Gao, M. Li, A. Bonti, W. Zhou, and S. Yu, “Multidimensionalrouting protocol in human-associated delay-tolerant networks,” MobileComputing, IEEE Transactions on, vol. 12, no. 11, pp. 2132–2144, Nov2013.