Learning-Based Cognitive Hitless Spectrum Defragmentation for Dynamic Provisioning in Elastic Optical Networks

Li, Ruoxing; Gu, Rentao; Jin, Weiqi; Ji, Yuefeng

doi:10.1109/lcomm.2021.3053279

Cited by 20 publications

(5 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The study in [20] highlighted DRL as a competitive alternative to established and well-known solutions when it comes to optimization problems in optical networks, e.g., Routing and Wavelength Assignment (RWA). A recent study in [21] applied DRL to solve the on-demand, reactive hitless SD problem. Upon an unsuccessful RMSA attempt, a DRL agent selects one of the pre-defined stretch schemes that extends the size of the fragmented spectrum to accommodate for blocked services.…”

Section: Related Workmentioning

confidence: 99%

DeepDefrag: a deep reinforcement learning framework for spectrum defragmentation

Etezadi¹,

Natalino²,

Diaz³

et al. 2022

Preprint

View full text Add to dashboard Cite

<p>An exponential growth of bandwidth demand, spurred by emerging network services, often with diverse characteristics and stringent performance requirements, drive the need for more dynamic operation of optical networks, efficient use of spectral resources, and automation. Spectrum fragmentation is one of the main challenges of dynamic, resource-efficient Elastic Optical Networks (EONs). Fragmented, stranded spectrum slots lead to poor resource utilization and increase the blocking probability of incoming service requests. Conventional approaches for Spectrum Defragmentation (SD) apply various criteria to decide when, and which portion of the spectrum to defragment. However, these polices often address only a subset of tasks related to defragmentation, are not adaptable, and have limited automation potential. To address these issues, we propose DeepDefrag, a novel framework based on reinforcement learning that addresses the main aspects of the SD process: determining when to perform defragmentation, which connections to reconfigure, and which part of the spectrum to reallocate them to. DeepDefrag outperforms the well-known Oldest-First FirstFit (OF-FF) defragmentation heuristic, substantially reducing blocking probability and defragmentation overhead.</p>

show abstract

Section: Related Workmentioning

confidence: 99%

DeepDefrag: a deep reinforcement learning framework for spectrum defragmentation

Etezadi¹,

Natalino²,

Diaz³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…In dynamic scenarios, DRL was applied to solve the routing, modulation, and spectrum assignment (RMSA) problem in single-domain EONs [27,28,30,31], multidomain EONs [32], multiband EONs [33,34] ,and survivable EONs operating under shared protection [35]; the problem of energy-efcient trafc grooming in fog-cloud EONs [36], the problem of establishing and reconfguring multicast sessions in EONs [37], the fragmentation mitigation problem [38], and the resource allocation problem with advanced reservation (AR) in EONs for cloud-edge computing [39]. Only one previous work has studied the application of DRL on MCF networks [40], but this work focused on fxed-grid networks.…”

Section: Related Workmentioning

confidence: 99%

Resource Allocation in Multicore Elastic Optical Networks: A Deep Reinforcement Learning Approach

et al. 2023

View full text Add to dashboard Cite

A deep reinforcement learning (DRL) approach is applied, for the first time, to solve the routing, modulation, spectrum, and core allocation (RMSCA) problem in dynamic multicore fiber elastic optical networks (MCF-EONs). To do so, a new environment was designed and implemented to emulate the operation of MCF-EONs - taking into account the modulation format-dependent reach and intercore crosstalk (XT) - and four DRL agents were trained to solve the RMSCA problem. The blocking performance of the trained agents was compared through simulation to 3 baselines RMSCA heuristics. Results obtained for the NSFNet and COST239 network topologies under different traffic loads show that the best-performing agent achieves, on average, up to a four-times decrease in blocking probability with respect to the best-performing baseline heuristic method.

show abstract

“…The authors in [34] investigate the problem of global optimization of network performance in a survivable EON use case and propose a DRL-based algorithm with the objective of improving the overall network performance in terms of cost value and survivability, where two RL agents are utilized to provide working and protection paths. In [35], DRL is used to tackle the on-demand, reactive hitless SD problem. Upon a failure of an incoming service request, the DRL agent selects one of the pre-defined schemes that increase the size of the fragmented spectrum to accommodate blocked services.…”

Section: Reinforcement Learning In Optical Networkmentioning

confidence: 99%

Deep reinforcement learning for proactive spectrum defragmentation in elastic optical networks

Etezadi,

Natalino,

Diaz

et al. 2023

J. Opt. Commun. Netw.

View full text Add to dashboard Cite

The immense growth of Internet traffic calls for advanced techniques to enable the dynamic operation of optical networks, efficient use of spectral resources, and automation. In this paper, we investigate the proactive spectrum defragmentation (SD) problem in elastic optical networks and propose a novel deep reinforcement learning-based framework DeepDefrag to increase spectral usage efficiency. Unlike the conventional, often threshold-based heuristic algorithms that address a subset of the defragmentation-related tasks and have limited automation capabilities, DeepDefrag jointly addresses the three main aspects of the SD process: determining when to perform defragmentation, which connections to reconfigure, and which part of the spectrum to reallocate them to. By considering service attributes, the spectrum occupancy state expressed by several different fragmentation metrics, and the reconfiguration cost, DeepDefrag is able to consistently select appropriate reconfiguration actions over the network lifetime and adapt to changing conditions. Extensive simulation results reveal superior performance of the proposed scheme over a scenario with exhaustive defragmentation and a well-known benchmark heuristic from the literature, achieving lower blocking probability at a smaller defragmentation overhead.

show abstract

Learning-Based Cognitive Hitless Spectrum Defragmentation for Dynamic Provisioning in Elastic Optical Networks

Cited by 20 publications

References 18 publications

DeepDefrag: a deep reinforcement learning framework for spectrum defragmentation

DeepDefrag: a deep reinforcement learning framework for spectrum defragmentation

Resource Allocation in Multicore Elastic Optical Networks: A Deep Reinforcement Learning Approach

Deep reinforcement learning for proactive spectrum defragmentation in elastic optical networks

Contact Info

Product

Resources

About