Title: Load-balancing contention resolution in OBS networks based on GMPLS
Authors: Haibo Wen, Huan Song, Lemin Li, Sheng Wang
Addresses: Key Lab of Broadband Optical Fiber Transmission and Communication Networks, University of Electronic Science and Technology of China, Chengdu 610054, China; Research and Innovation Center, Alcatel Shanghai Bell, Shanghai 201206, China. ' Key Lab of Broadband Optical Fiber Transmission and Communication Networks, University of Electronic Science and Technology of China, Chengdu 610054, China; Networking Lab, University of California, Davis CA, USA. ' Key Lab of Broadband Optical Fiber Transmission and Communication Networks, University of Electronic Science and Technology of China, Chengdu 610054, China. ' Key Lab of Broadband Optical Fiber Transmission and Communication Networks, University of Electronic Science and Technology of China, Chengdu 610054, China
Abstract: Contention resolution (CR) in optical burst switching (OBS) networks is a hot topic in the academia. However, most previous research was done within the local core switching node. In this paper, we investigate an integrated CR technique in OBS networks based on generalised multi-protocol label switching (GMPLS). At first, a new network model is proposed, in which a label switched path (LSP) is just composed of a series of links and not assigned with any wavelength, and the wavelengths are decided when core switches process the control packets. Then, we propose a novel technique, i.e., load-balancing contention resolution (LBCR), which is divided into two parts: within the network layer, CR is done with load-balancing technology; within the node layer, local CR algorithms can be employed. Simulations show that LBCR can reduce burst contention and burst loss probability.
Keywords: optical burst switching; contention resolution; wavelength division multiplexing; routing assignment; wavelength assignment; generalised multi-protocol label switching; GMPLS; OBS networks; networking; high performance computing; load balancing; loss probability; simulation.
DOI: 10.1504/IJHPCN.2005.007864
International Journal of High Performance Computing and Networking, 2005 Vol.3 No.1, pp.25 - 32
Published online: 28 Sep 2005 *
Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article