Future works

The work presented in this thesis opens the ways to several directions for future work. As future work, the formulation and performance evaluation of both PA and LD models under traffic uncertainty as well as including layers such as the optical one for the LD model shall be considered.

In this thesis we considered a situation where traffic demand is known but in actual commercial network traffic is uncertain [8, 10, 12, 37, 52, 55]. In fact traffic highly varies from day to night in the suburbs of big cities as people move from home to their work place. It also changes due to more unpredictable factors such as disasters [13, 43, 47], software updates, games or movie releases, invents and so on. These factors tremendously increase traffic demand at a specific place and time [26, 45, 69]. In the basic model of link weight optimization, traffic uncertainty was addressed [63, 64]. As a next step traffic uncertainty in the PA model and its related schemes shall be considered. In the LD model too, traffic unpredictability should be considered when pinpointing the links that need to be duplicated.

Moreover, the LD model and its dependent schemes can be expanded to in-clude the optical layer as we only focused on the data link layer and the IP layer for link duplication. By considering the optical layer, additional resources can be inspected at a lower granularity since bundle of optical links constitute a link a the data link layer. This may lead to a reduction of the required additional resources as added discrete resources become less sparce and resource sharing among links of the same bundle can envisaged.

However at the optical layer, shared risk link groups (SRGs) [44, 57] shall be taken into consideration [49, 56, 58]. SRG is a group of link that share a common resource or function such that when malfunction occurs all the related links fail. In the optical layer bundled links are grouped together in a span [1].

This increases the risk of having all links failed if a span cut occurs. This kind of event may happen during disasters and large scale network attacks and affects

7.2 Future works

network operation [14, 16, 24, 39]. PA and LD models should be upgraded to also deal with simultaneous multi-link failures as this thesis addresses only single link failures. The work in [28] considers an optimization problem to reduce the backup resources under random and simultaneous link failures by introducing the problematic survivability. PA and LD models should be extended to deal with random and simultaneous link failures to keep the problematic survivability.

Furthermore, disasters and network attacks are less likely to happen compare to traffic spikes or software updates [7, 21, 59]. In that sense simultaneous link failures rarely happen. A permanent protection against these events will result in costly networks in an era where data related revenue is decreasing with the spread of Mobile Virtual Network Operators (MVNOs) [5, 11, 20, 35, 40]. There is a need to include probabilistic factors [57] when designing new PA and LD models in order to deal with random and simultaneous link failures at a lower cost.

Finally, PA and LD models should be investigated and compared in the frame-work of netframe-work function virtualization (NFV) and software defined netframe-working (SDN). NFV decouples network functions from proprietary hardware to enable network operators to easily scale resources on demand at any given location above a cheaper hardware. SDN makes it possible to dynamically set or update commu-nication paths based on specific predefined events [66]. In this thesis we did not consider the factors that NFV and SDN bring to the table in terms of network traffic engineering. Since NFV/SDN is gradually being introduced in commer-cial networks [6], it should be taken into consideration when investigating the applicability of PA and LD models.

7. CONCLUSIONS AND FUTURE WORKS

Bibliography

[1] Flex-span adss fiber optic cable. https://www.aflglobal.com/Products/. 68 [2] The internet2 network. http://www.internet2.edu/network/. 29

[3] Ip routing concepts and the process of next-hop routing.

http://www.tcpipguide.com/free/. 1

[4] Reducing link failure and topology change notification times in is-is networks.

http://www.cisco.com/en/US/docs/ios-xml/ios/iproute isis/. 4

[5] Paolini: Is cost per bit a reliable metric for wireless data?

http://www.fiercewireless.com/tech/, 2011. 69

[6] Docomo develops first nfv technology for multi-vendor epc software.

https://www.nttdocomo.co.jp/english/info/media center/pr/2016/0219 00.html, 2016. 69

[7] O. Alhazmi and Y. K. Malaiya. Evaluating disaster recovery plans using the cloud. 2013. 69

[8] D. Applegate and E. Cohen. Making intra-domain routing robust to changing and uncertain traffic demands: understanding fundamental tradeoffs. IEEE, 2003. 68

[9] A. Arora. Dijkstra’s algorithm. http://www.slideshare.net/ami 01/, 2011. 3 [10] Towles B. and W.J. Dally. Worst-case traffic for oblivous routing functions.

IEEE Computer Architecture Letters, 1(1), 2002. 68

BIBLIOGRAPHY

[11] M. Balon and B. Liau. Mobile virtual network operator. In Inter-national Telecommunications Network Strategy and Planning Symposium (NEXTWORKS), 2012. 69

[12] M. Bienkowski, M. Korzeniowski, and H. Racke. A practical algorithm for constructing oblivious routing schemes. IEEE, 2003. 68

[13] R. Biswas, Bhattacharyya D., and S.K. Bandyopadhyay. Command and control (c2) network for disaster management. IEEE, 2012. 68

[14] E. Bouilllet, G. Ellinas, J.R. Labourdette, and R. Ramamurthy. Path routing in mesh optical networks. 69

[15] L. S. Buriol, M. C. Resende, C. C. Ribeiro, and M. Thorup. A hybrid genetic algorithm for the weight setting problem in ospf-is-is routing. Networks, 46(1):36–56, 2005. 3

[16] H. Choi, S. Subramaniam, and H. Choi. On double-link failure recovery in wdm optical networks. volume 2, pages 808–816. IEEE, 2002. 69

[17] J. Chu and C. Lea. Optimal link weights for maximizing qos traffic. InIEEE International Conference on Communications, pages 610–615. IEEE, 2007.

29

[18] J. Chu and C. Lea. Optimal link weights for ip-based networks supporting hose-model vpns. IEEE/ACM Transactions on Networking, 17(3):778–788, 2009. 13, 15

[19] M. Clouqueur and W.D. Grover. Availability analysis of span-restorable mesh networks. ‘IEEE Journal on Selected Areas Communications, 20(4):810–821, 2002. 5

[20] M. Debbah, L. Echabbi, and C. Hamlaoui. Market share analysis be-tween mno and mvno under brand appeal based segmentation. In Inter-national Conference on Network Games, Control and Optimization (NetG-Coop), 2012. 69

BIBLIOGRAPHY

[21] S. Ferdousi, F. Dikbiyik, M.F. Habib, M. Tornatore, and B. Mukherjee.

Latest trends in traffic matrix modeling and its application to multilayer te.

Journal of optical communications and networking, 7(7):681–694, 2015. 69

[22] B. Fortz, J Rexford, and M. Thorup. Traffic engineering with traditional ip protocols. IEEE Communication Magazine, 40:759–767, 2002. 3, 25

[23] B. Fortz and M. Thorup. Optimizing ospf/is-is weights in a changing world.

IEEE Journal on Selected Areas in Communications, pages 759–767, 2002.

3, 25

[24] L. Guo, H. Yu, and L. Li. Segment shared protection for survivable meshed wdm optical networks. Optical Communications, 251(4–6):328–338, 2005. 69 [25] Qiu. Hangping, Lei. Zhipeng, and Huang Haiyan. An evaluation method for qoe of information service based on two-tuple linguistic. In6th International Conference on Intelligent Systems Design and Engineering Applications (IS-DEA). IEEE, 2016. 45

[26] R. Hayashi, T. Miyamura, D. Shimazaki, E Oki, and K. Shiomoto. Disaster-aware datacenter placement and dynamic content management in cloud net-works. IEICE Transactions on Communications, E90-B(8):1912–1921, 2007.

68

[27] G. Iannaccone, C. Chuah, R. Mortier, S. Bhattacharyya, and C. Diot. Anal-ysis of link failures in a large ip backbone. In 2nd ACM SIGCOM Internet Measurement Workshop, pages 237–242. ACM/SIGCOM, 2002. 4, 5, 11

[28] M. Johnston, H.W. Lee, and E. Modiano. A robust optimization approach to backup network design with random failures. IEEE/ACM Transactions on Networking, DOI: 10.1109_/TNET.2014.2320829, pages 1216–1228, 2015.

69

[29] I. M. Kamrul and E. Oki. Pso: Preventive start-time optimization of ospf link weights to counter network failure. IEEE Communications Letters, 14(6):581–583, 2010. 4, 5, 15, 25

BIBLIOGRAPHY

[30] I. M. Kamrul and E. Oki. Optimization of ospf link weight to minimize worst-case network congestion against single-link failure. InIEEE ICC, pages 1–5.

IEEE, 2011. 4, 22, 32, 44, 53, 63, 65

[31] S. Kaptchouang, Ou´edraogo I. A., and E. Oki. Preventive start-time op-timization of link weights with link reinforcement. IEEE Communications Letters, 18(7):1179–1182, 2014. iv, 7, 44, 45, 47

[32] S. Kaptchouang, Ou´edraogo I. A., and E. Oki. Strengthened preventive start-time optimisation that reduces congestion ratio under no failure scenario.

IET, The Journal of Engineering, DOI: 10.1049/joe.2016.0179, 2016. (to appear), 2016. 6, 19, 38

[33] S. Kaptchouang, Tahara H., and E. Oki. Link weight optimization scheme for link reinforcement in ip networks. IEICE Transactions on Communications., E100-B(3), 2017. iv, 7, 49, 51, 53, 55

[34] S. Kaptchouang and E. Oki. Enhancing preventive start-time optimization considering both failure and non-failure scenario. In 19th Asia-Pacific Con-ference on Communications, pages 763–766. IEEE, 2013. 6, 19

[35] B.W. Kim, C.Y. Ko, and S.A. Kang. Data mvno: Cost-based pricing in korea. In Technology Management for Emerging Technologies (PICMET), 2012. 69

[36] M Kodialam and T.V. Lakshman. Dynamic routing of bandwidth guaranteed tunnels using aggregated network resource usage information. IEEE/ACM Transactions on Network, 11(3):399–410, 2003. 5

[37] M. Kodialam, T.V. Lakshman, J.B. Orlin, and S. Sengupta. Oblivious rout-ing of highly variable traffic in service overlays and ip backbones.IEEE/ACM Transactions on Networking, 17(2):459–472, 2008. 68

[38] Y. Liu, D. Tipper, and P. Siripongwutikorn. Approximating optimal spare capacity allocation by successive survivable routing. IEEE/ACM TRANS-ACTIONS ON NETWORKING, 13(1):198–211, 2005. 11

BIBLIOGRAPHY

[39] S. et al Lumetta. Capacity versus robustness: A tradeoff for link restoration in mesh networks. Journal Lightwave Technology, 18(12):1765, 2000. 69 [40] C.M. Machuca, T. Krauss, A. Basta, E. Grigoreva, and W. Kellerer.

Alter-natives for the automotive industry to become an mvno. In International Conference on Transparent Optical Networks (ICTON), 2014. 69

[41] A. Markopolou, G. Iannaccone, C. Chuah, S. Bhattacharyya, and C. Diot.

Characterization of link failures in ip backbone. In IEEE INFOCOM, vol-ume 4, pages 2307–2317. IEEE, 2004. 4, 11

[42] J. Martin. How to connect to your router to change settings.

http://www.pcadvisor.co.uk/how-to/network-wifi/, 2016. 2

[43] C. Maru, M. Enoki, A. Nakao, S. Yamamoto, S. Yamaguchi, and M. Oguchi.

Network failure detection system for traffic control using social information in large-scale disasters. IEEE, 2015. 68

[44] J. Matthew, W. L. Hyang, and M. Eytan. A robust optimization approach to backup network design with random failures.IEEE/ACM TRANSACTIONS ON NETWORKING, 23(4), 2015. 68

[45] A. Medina, N. Taft, K. Salamatian, R., S. Bhattacharyya, and C. Diot.

Traffic matrix estimation: Existing techniques and new directions. InACM SIGCOM, pages 161–174. ACM/SIGCOM, 2002. 68

[46] Alberto Medina, Anukool Lakhina, Ibrahim Matta, and John Byers. Brite:

An approach to universal topology generation. In Proceedings. Ninth Inter-national Symposium on Modeling, Analysis and Simulation of Computer and Telecommunication Systems, 2001., pages 346–353. IEEE, 2001. 29

[47] Alberto Medina, Anukool Lakhina, Ibrahim Matta, and John Byers. Network adaptability from disaster disruptions and cascading failures. In Ninth In-ternational Symposium on Modeling, Analysis and Simulation of Computer and Telecommunication Systems, 2001., pages 346–353. IEEE, 2001. 68

[48] J. T. Moy. Ospf version 2. IETF RFC 1247, 1991. 2

BIBLIOGRAPHY

[49] S. Okamoto, E. Oki, K. Shimano, A. Sahara, and N. Yamanaka. Demonstra-tion of the highly reliable hikari router network based on a newly developed disjoint path selection scheme. IEEE Communications Magazine, 40(11):52–

59, 2002. 68

[50] E. Oki. Linear programming and algorithms for communication networks.

2012. 53

[51] E. Oki. Preventive start-time optimization against network failure. In 10th Conference on Design of Reliable Communication Networks. IEEE, 2014. 6

[52] E. Oki and A. Awaki. Fine two-phase routing over shortest paths with traffic matrix. Computer Networks, 54(18):3223–3231, 2010. 68

[53] E. Oki and A. Awaki. Load-balanced ip routing scheme based on shortest paths in hose model. IEEE Transactions on Communications, 58(7):2088–

2096, 2010. 3, 12

[54] E. Oki and R. R. Cessa. Ip routing protocols. Advanced Internet Protocols, Services, and Applications, pages 39–40, 2012. v, 2

[55] E. Oki, Y. Kitahara, and I. A. Ou´edraogo. Optimal routing by hose model with bound of link traffic. IET Networks, 1(3):155–162, 2012. 68

[56] E. Oki, J.L. Le Roux, and A. Farrel. Framework for pce-based inter-layer mpls and gmpls traffic engineering. IETF RFC 5623, 2009. 68

[57] E. Oki, N. Matsuura, K. Shiomoto, and N. Yamanaka. A disjoint path selection scheme with shared risk link groups in gmpls networks. IEEE Communications Letters, 6(9), 2002. 68, 69

[58] E. Oki, T. Takeda, and A. Farrel. Extensions to the path computation element communication protocol for route exclusions.IETF RFC 5521, 2009.

68

[59] L. Pang, J. Zhou, and D. Liu. The probability prediction of storm surge disaster along south-east coasts of china. IEEE, 2011. 69

BIBLIOGRAPHY

[60] M. Pirlo, A. Szentesi, J. Harmatos, A. Jttner, P. Gajowniczek, and S. Kozdrowski. On ospf related network optimization problems. Perfor-mance Evaluation, 48:201–223, 2002. 15

[61] A. Popescu. Dealing with qoe and power consumption in video distribution networks. InInternational Conference on Communications (COMM). IEEE, 2016. 45

[62] S. Ramamurthy, L. Sahasrabuddhe, and B. Mukherjee. Survivable wdm mesh. IEEE Networks, 21(4):870–883, 2003. 11

[63] R. S. Ranaweera, I. M. Kamrul, and E. Oki. Preventive start-time opti-mization of ospf link weights against link failure for hose model. In 18th Asia-Pacific Conference on Communications, pages 698–702. IEEE, 2012.

4, 68

[64] R. S. Ranaweera, I. M. Kamrul, and E. Oki. Preventive start-time optimiza-tion of ospf link weights for hose model. IET Networks, 2014. 4, 68

[65] C. Reichert and Magedanz T. A fast heuristic for genetic algorithms in link weight optimization. Lecture Notes in Computer Science, 3266, 2004. 3 [66] A. Sezer, S.C Hayward, P.K. Chouhan, B. Fraser, D. Lake, Finnegan J.,

N. Vijoen, M. Miller, and N. Rao. Are we ready for SDN? implementation challenges for software-defined networks. IEEE Communications magazine, 51(7):36–43, 2013. 69

[67] D. Thaler and Hopps C. Multipath issues in unicast and multicast next-hop selection. IETF RFC 2991, 2000. 12

[68] I. Wang and M. R. Ito. Dynamics of load sensitive adaptive routing. In IEEE International Conference on Communications, volume 1, pages 213–

217. IEEE, 2005. 3

[69] H. Weber. Mobile web traffic spikes 35% in a year, underlines the smartphone revolution. http://thenextweb.com/insider/2012/03/03/, 2012. 68

BIBLIOGRAPHY

[70] J. Zhang and B. Mukherjee. A review of fault management in wdm mesh networks: Basic concepts and research challenges. IEEE Networks, 18(2):41–

48, 2004. 11

Publications

List of Publications related to the thesis

Journal Papers

1. S. Kaptchouang, I.A. Ou´edraogo, and E. Oki, “Preventive Start-time Op-timization Considering Both Failure and Non-Failure Scenarios,” IEICE Transaction on Communications, vol. E100-B, no.7, ,Jul. 2017 (Accepted) 2. S. Kaptchouang, H. Tahara, and E. Oki, “Link Weight Optimization Scheme for Link Reinforcement in IP Networks.” IEICE Transaction on Communi-cations, vol. E100-B, no. 03, 2017. (Accepted)

3. S. Kaptchouang, I.A. Ou´edraogo, and E. Oki, “Strengthened Preventive Start-Time Optimisation that Reduces Congestion Ratio under No Failure Scenario,” The Journal of Engineering, DOI: 10.1049/joe.2016.0179, 2016.

(Accepted)

4. S. Kaptchouang, I.A. Ou´edraogo, and E. Oki, “Preventive Start-Time Opti-mization of Link Weights with Link Reinforcement,” IEEE Commun. Let., vol. 18, no. 7, pp. 1179-1182, Jul. 2014. (letter)

International Conference Papers

1. S. Kaptchouang and E. Oki, “Enhancing Preventive Start-Time Optimiza-tion Considering Both Failure and Non-Failure Scenarios,”19th Asia-Pacific Conference on Communications (APCC 2013), Aug. 2013.