UNCORRECTED PROOF OSN: 113 Model 3G pp. 1–21 (col. fig: nil) ARTICLE IN PRESS Optical Switching and Networking xx (xxxx) xxx–xxx Contents lists available at ScienceDirect Optical Switching and Networking journal homepage: www.elsevier.com/locate/osn Automated adaptive on-line Multi-Layer Traffic Engineering through ‘‘tailoring’’ wavelength-paths in the Fragment Graph Tibor Cinkler, Péter Hegyi ∗ High Speed Networks Laboratory, Department of Telecommunications and Media Informatics, Budapest University of Technology and Economics, Magyar tudósok körútja 2, H-1117 Budapest, Hungary article info Article history: Received 25 February 2008 Received in revised form 20 May 2009 Accepted 22 May 2009 Available online xxxx Keywords: Automated adaptive Multi-Layer Traffic Engineering FG Fragment Graph Wavelength path fragmentation and de-fragmentation Traffic grooming WDM Routing RWA abstract In Multi-Layer networks, where more than one layer is switched, i.e., connections are set up using not only the upper, e.g., IP layer but the underlying wavelength layer as well leads often to suboptimal performance due to long wavelength paths, that do not allow routing the traffic along the shortest path. The role of MLTE (Multi-Layer Traffic Engineering) is to cut (fragment) these long wavelength-paths into parts (fragments) that allow better routing at the upper layer, or to concatenate (defragment) two or more fragments into longer paths when the network load is low and therefore ∧ fewer hops are preferred. In this paper we present a new model, the Fragment Graph (FG) and an algorithm for this model that supports Fragmentation and De-Fragmentation of wavelength paths making the network always instantly adapt to changing traffic conditions. We introduce the notion of shadow links to model ‘‘λ-path tailoring’’. We implicitly assume that the wavelength paths carry such, e.g., IP traffic that can be interrupted for a few milliseconds and that even allows minor packet reordering. To show the superior performance of our approach in various network and traffic conditions we have carried out an intensive simulation study where we compare blocking ratios and path lengths and we analyze the dynamic behavior and fairness of the proposed and of reference methods. © 2009 Elsevier B.V. All rights reserved. 1. Introduction 1 The evolution of transport networks shows two main 2 directions. First, there are multiple networking technolo- 3 gies layered one over the other. Second, it is required that 4 not only the upper-most layer ∧ be switched, but the upper 5 two, or maybe all the layers. 6 Under the switched layer we assume that connections 7 can be set up on demand via user signaling and that 8 the network resources of multiple layers adapt these 9 requirements to best accommodate the new demands. To 10 emphasis is on multiple layer, e.g. a demand at the upper- 11 most layer can trigger setting up a path in a lower layer. 12 ∗ Corresponding author. Tel.: +36 14634391. E-mail addresses: cinkler@tmit.bme.hu (T. Cinkler), hegyi@tmit.bme.hu (P. Hegyi). Here we consider the case of Wavelength Routing 13 Dense Wavelength Division Multiplexing (WR-DWDM) 14 Networks and one layer built over it. WR-DWDM is a 15 technology applied in optical networks: in a fiber not 16 only one communication channel exists, but as many as 17 the number of wavelengths that can be demultiplexed 18 at the terminations of the fiber. In the WR-DWDM 19 layer a wavelength path (λ-path) connects two physically 20 adjacent or distant nodes. These two physical nodes will 21 seem adjacent for the upper layer built over it. 22 The upper layer is an ‘‘electronic’’ one. In switched 23 multi-layer transport networks (e.g. ASTN/GMPLS) the 24 traffic demands have typically bandwidth by orders of 25 magnitude lower than the capacity of λ-links. Therefore, 26 it is not worth assigning exclusive end-to-end λ-paths 27 to these demands, i.e., sub-λ granularity is required. 28 Furthermore, the number of λs per fiber is limited and 29 expensive. To increase the throughput of a network with 30 1573-4277/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.osn.2009.05.002 Please cite this article in press as: T. Cinkler, P. Hegyi, Automated adaptive on-line Multi-Layer Traffic Engineering through ‘‘tailoring’’ wavelength-paths in the Fragment Graph, Optical Switching and Networking (2009), doi:10.1016/j.osn.2009.05.002