A First Step Towards Autonomic Optical Burst Switched Networks

Praveen, J.; Praveen, B.; Murthy, C. Siva Ram

doi:10.1109/icac.2005.1

Cited by 5 publications

(3 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…2 and Fig. 3 show the average BLP of QRWA algorithm with increasing load compared to that for Linear Reward Penalty (LRP)-based algorithm proposed in [8], shortest path first with random wavelength assignment(SPF+RW), SPF with destination priority variant (SPF+PWA(Dest)) proposed in [5], SPF with link priority variant (SPF+PWA(Link)) proposed in [6] and a cooperation based RWA algorithm (Compound) proposed in [9] for the random topology and the NSFNET topology, respectively. The results are presented for the dynamic load to demonstrate the adaptability of the QRWA algorithm to changes in the network conditions.…”

Section: Simulation Resultsmentioning

confidence: 95%

“…It is known that any algorithm that learns an optimal policy over the time through exploration outperforms the schemes based on fixed policy [7]. The authors in [8] proposed algorithms based on learning automata that select route and wavelength for each burst. Algorithms to minimize the BLP based on Q-learning were proposed for path selection and wavelength selection in [7].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Joint Path and Wavelength Selection Using Q-learning in Optical Burst Switching Networks

Venkatesh

Kiran

Murthy

2009

2009 IEEE International Conference on Communications

View full text Add to dashboard Cite

Contention losses which usually do not indicate congestion is a major issue that hinders the deployment of optical burst switching (OBS) networks. Development of efficient path and wavelength selection algorithms is crucial to minimize the burst loss probability (BLP) in OBS networks. In this paper, we handle path selection and wavelength selection in a joint fashion. We formulate the problem of selecting a pair of path and wavelength jointly as a multi-armed bandit problem (MABP) and discuss the difficulties in solving MABP directly. We then rewrite the Q-learning formalism to solve the MABP without explicit model in an online fashion and propose an algorithm to solve the problem near-optimally. The proposed algorithm selects a pair of path and wavelength at each ingress node to minimize the BLP on the long run. Simulation results demonstrate the effectiveness of our algorithm in minimizing the BLP with better link utilization compared to the other proposals in the literature.

show abstract

Section: Simulation Resultsmentioning

confidence: 95%

Section: Introductionmentioning

confidence: 99%

Joint Path and Wavelength Selection Using Q-learning in Optical Burst Switching Networks

Venkatesh

Kiran

Murthy

2009

2009 IEEE International Conference on Communications

View full text Add to dashboard Cite

show abstract

“…We use a technique similar to the one used in [12] to compute the link loss porbabilities of the individual links. We assume that R-candidate routes to every destination are pre-computed and are available at all source nodes.…”

Section: Link and Route Loss Probability Estimationmentioning

confidence: 99%

Reducing Data Loss in Optical Burst-Switched Networks Using Adaptive Burst Cloning

Rezaie

Motlagh

Khorsandi³

et al. 2007

15th International Conference on Advanced Computing and Communications (ADCOM 2007)

View full text Add to dashboard Cite

In this paper, we introduce a loss reduction scheme, Adaptive Burst cloning, as a technique to ensure service-guarantees in optical burst-switched (OBS) networks. The basic idea is to replicate a burst and send its duplicate copies through the network in parallel. Incase the original burst gets lost, the cloned burst may still be able to reach the destination. Unlike the existing burst cloning techniques, Adaptive Burst Cloning selects the clone node and the number of cloned bursts dynamically. Due to dynamic selection of the clone node and the number of cloned bursts, utilization of network resources is considerably low, and as a consequence, burst loss due to resource contention is significantly reduced. We develop a network level simulation model to investigate the proposed Adaptive Burst Cloning loss recovery mechanism and compare its performance with the existing static burst cloning technique. Our results show that the proposed mechanism significantly reduces the packet loss in an optical burst-switched network with relatively lesser resource utilization.

show abstract