Evaluation of Topology Optimization Objectives in IP Networks

다국어 초록 (Multilingual Abstract)

In the past, various optimization objective functions havebeen proposed to help in network optimization, especially for usein traffic engineering (TE) and topology optimization. This varietyof optimization objectives resulted in the emergence of algorithmstargeting different objectives. However, the role of the objectivefunction has been largely overlooked. Because, the choiceof a particular objective function was not justified in most of thecases. Some researchers criticized this arbitrary selection of objectivefunctions. Even though some researchers intuitively suggestusing a specific objective, only few work tackled with the problemof evaluating the objectives. In this paper, we evaluate various networkoptimization objectives on topology optimization. Previously,a study analyzed the efficiency of some routing optimization objectivesusing linear programming (LP) by linear relaxation. However,some of the objective functions are nonlinear, and such a linear relaxationdoes not treat each objective equally.The difficulty arisesdue to the fact that optimization algorithms are objective functiontailored heuristics. To achieve fairness, we compare and analyzedifferent traffic optimization objectives for topology optimizationusing neural networks which are used to model nonlinear relations.
By using neural networks, we strive to avoid any unfairness, suchas obviating linear approximation. Also, our work suggests whichfeatures are meaningful for machine learning in network optimization.
Our method partially agrees with the previous work, and weconclude that delay is the best performing optimization objective.

참고문헌 (Reference)

1 A. K. Dutta, "WDM technologies : Optical networks. Volume III" Elsevier Academic Press 2004

2 S. Agarwal, "Traffic engineering in software defined networks" 2211-2219, 2013

3 A. Nucci, "The problem of synthetically generating IP traffic matrices : initial recommendations" 35 (35): 19-32, 2005

4 D. O. Hebb, "The organization of behaviour" Wiley 1949

5 P. Borgnat, "Seven years and one day: Sketching the evolution of internet traffic" 711-719, 2009

6 W. B. Gong, "Self-similarity and long range dependence on the internet : A second look at the evidence, origins and implications" 48 (48): 377-399, 2005

7 D. O. Awduche, "RFC 2702: Requirements for traffic engineering over MPLS"

8 S. Uhlig, "Providing public intradomain traffic matrices to the research community" 36 (36): 83-, 2006

9 P. Sangkatsanee, "Practical real-time intrusion detection using machine learning approaches" 34 (34): 2227-2235, 2011

10 B. Fortz, "Optimizing OSPF/IS-IS weights in a changing world" 20 (20): 756-767, 2002

11 J. Zheng, "Optical WDM Networks: Concepts and design principles" John Wiley & Sons 2004

12 W. E. Leland, "On the self-similar nature of Ethernet traffic(extended version)" 2 (2): 1-15, 1994

13 O. Heckmann, "On realistic network topologies for simulation" 28-32, 2003

14 R. Rojas, "Neural networks: A systematic introduction" Springer-Verlag 1996

15 K. Kar, "Minimum interference routing of bandwidth guaranteed tunnels with MPLS traffic engineering applications" 18 (18): 2566-2579, 2000

16 A. Elwalid, "MATE : Multipath adaptive traffic engineering" 40 (40): 695-709, 2002

17 D. Levin, "Logically centralized? state distribution trade-offs in software defined networks" 1-, 2012

18 I. Chlamtac, "Lightpath communications : An approach to high bandwidth opticalWAN’s" 40 (40): 1171-1182, 1992

19 N. Degrande, "Inter-area traffic engineering in a differentiated services network" 11 : 427-445, 2003

20 S. Balon, "How well do traffic engineering objective functions meet TE requirements" 3976 : 75-86, 2006

21 Y. S. Hanay, "Evaluation of topology optimization objectives" 458-461, 2015

22 J. He, "Don’t optimize existing protocols, design optimizable protocols" 37 (37): 53-58, 2007

23 G. Tanaka, "Complex-valued multistate associative memory with nonlinear multilevel functions for gray-level image reconstruction" 20 (20): 1463-1473, 2009

24 N. N. Aizenberg, "CNN based on multi-valued neuron as a model of associative memory for grey scale images" 36-41, 1992

25 CISCO, "Best practices in core network capacity planning architectural principles of the MATE portfolio of products"

26 N. Wang, "An overview of routing optimization for internet traffic engineering" 10 (10): 36-56, 2008

27 A. Sivaraman, "An experimental study of the learnability of congestion control" 479-490, 2014

28 F. Blanchy, "An efficient decentralized on-line traffic engineering algorithm for MPLS networks" 2003

29 Y. Koizumi, "Adaptive virtual network topology control based on attractor selection" 28 (28): 1720-1731, 2010

30 A. Balachandran, "A quest for an internet video quality-of-experience metric" 97-102, 2012

31 T. Karagiannis, "A nonstationary Poisson view of Internet traffic" 1558-1569, 2004

32 Y. Jin, "A Modular machine learning system for flow-level traffic classification in large networks" 6 (6): 1-34, 2012

33 A. Altin, ""Intra-domain traffic engineering with shortest path routing protocols" 204 (204): 2013