Knowledge-Based Autonomous Service Provisioning in Multi-Domain Elastic Optical Networks

Chen, Xiaoliang; Proietti, Roberto; Lu, Hanchao; Castro, Alberto; Yoo, Sung Jong

doi:10.1109/mcom.2018.1701191

Cited by 52 publications

(38 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…. These values are used because the BER at the receiver nodes is known only for the already established connections (through OPM [5,6]).…”

Section: Deep Graph Convolutional Neural Network For Qot Estimationmentioning

confidence: 99%

“…1, where an optical network is centrally controlled by an SDN-based controller [5] that dynamically monitors and configures the data plane. A database platform collects and stores real-time network state information and optical network monitoring information through OPM [5,6]. Specifically, it collects and stores information capable of fully describing any network state, G s , and its ground truth y s .…”

Section: Data-driven Qot Frameworkmentioning

confidence: 99%

“…Among the ML methods applied (K-nearest neighbors, logistic regression, support vector machines, and neural networks (NNs)), NNs have shown to present better generalization and higher accuracy [4]. NNs' accuracy is also experimentally demonstrated in [5], while a software defined networking (SDN)-based implementation scheme incorporating the data-driven QoT estimation framework is presented in [6].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Machine Learning for QoT Estimation of Unseen Optical Network States

Panayiotou¹,

Savva²,

Shariati³

et al. 2019

Optical Fiber Communication Conference (OFC) 2019

View full text Add to dashboard Cite

show abstract

“…. These values are used because the BER at the receiver nodes is known only for the already established connections (through OPM [5,6]).…”

Section: Deep Graph Convolutional Neural Network For Qot Estimationmentioning

confidence: 99%

Section: Data-driven Qot Frameworkmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Machine Learning for QoT Estimation of Unseen Optical Network States

Panayiotou¹,

Savva²,

Shariati³

et al. 2019

Optical Fiber Communication Conference (OFC) 2019

View full text Add to dashboard Cite

show abstract

“…ML applications for optical network planning have in the past few years attracted significant attention [9,10]. In their general form, existing ML applications assume an SDN-based optical network controller that centrally controls and manages the network [11,36] (Fig. 1).…”

Section: Related Workmentioning

confidence: 99%

“…This is especially true in current networks, where network planning functions are becoming increasingly complex in an uncertain network environment that is continuously changing, supporting heterogeneous applications and services. Existing ML applications [9,10] focus on traffic demand predictions and resource allocation optimization [11][12][13][14][15], fault detection/localization [16][17][18][19], attack detection/identification [20,21], and quality-of-transmission (QoT) estimation [22][23][24][25][26]. In most of these works, however, the diverse optical service level agreements (OSLAs) of the next generation optical networks [27] are not specifically considered.…”

mentioning

confidence: 99%

Decentralizing machine-learning-based QoT estimation for sliceable optical networks

Panayiotou

Savva

Tomkos

et al. 2020

J. Opt. Commun. Netw.

View full text Add to dashboard Cite

Dynamic network slicing has emerged as a promising and fundamental framework for meeting 5G's diverse use cases. As machine learning (ML) is expected to play a pivotal role in the efficient control and management of these networks, in this work, we examine the ML-based quality-of-transmission (QoT) estimation problem under the dynamic network slicing context, where each slice has to meet a different QoT requirement. Specifically, we examine ML-based QoT frameworks with the aim of finding QoT model/s that are fine-tuned according to the diverse QoT requirements. Centralized and distributed frameworks are examined and compared according to their model accuracy, routing and spectrum allocation (RSA) accuracy, and CPU (training time) and RAM (memory) requirements. We show that the distributed QoT models outperform the centralized QoT model in accuracy and CPU usage. The RSA accuracy, i.e., measuring the accuracy of the models with regard to the QoT-aware RSA decisions, is sufficiently high for both frameworks. Regarding the RAM usage, as the distributed framework has to train in parallel several QoT models, it may require higher memory, especially as the number of diverse QoT requirements increases. This memory, however, tends to be reserved for a shorter period of time. Moreover, this work develops a dynamic multi-slice QoT-aware (RSA) framework that integrates the ML-based QoT models. The aim is to examine the network performance when the diverse QoT models are considered, as opposed to the state-of-the-art single-slice QoT-aware RSA approach where all connections/slices are provisioned according to a single QoT requirement. We show that the multi-slice QoT-aware RSA approach significantly improves network performance, a clear indicator that the commonly considered single-slice QoT-aware RSA approach may lead to connection overprovisioning.few years adopted network slicing as an emerging technology and business model [2,3].In general, the implementation of network slicing is conceived as an end-to-end feature spanning over different technology domains. While relevant work focuses mostly on the implementation of network slicing in the radio access network (RAN) segment [1,2,4,5], recent work also focuses on extending the scope of network slicing beyond the RAN to include both mobile and transport (optical) network segments [6][7][8]. In particular, recent work deals with SDN/NFV demonstrations, network orchestration implementations, and traffic demand analysis, towards the vision of end-to-end network slicing. In general, SDN/NFV, network orchestration, and analytics are considered as the key ingredients for successful implementation of end-to-end network slicing [1]. However, regarding the efficient control and management of these slices (i.e., enabling not only their creation but also their 1943-0620/20/070146-17 Journal smart deployment at the right place, at the right time, with optimized resources), little has been done so far.Without doubt, machine learning (ML) will be instrumental in the efficient control an...

show abstract

The intelligent multi‐domain service function chain deployment: Architecture, challenges and solutions

Zhang

Wang

Dong

et al. 2020

Int J Communication

View full text Add to dashboard Cite

Network function virtualization (NFV) technology achieves flexible service deployment by replacing the middleboxes with virtual network functions (VNFs). In NFV, a set of VNFs are chained in a given order, called service function chain (SFC), and accordingly, data flow is steered to traverse all the VNFs in order to offer a service. With a large number of network devices and end users being connected into Internet, there is a growing demand for largescale multi-domain networks to dynamically deploy the SFC across multiple network domains, in order to support efficient service provisioning. To this end, in this paper, we first investigate the state of the art of multi-domain SFC deployment, and then propose an intelligent multi-domain SFC deployment (IMSD) architecture by leveraging software-defined networking (SDN), NFV, and deep learning technologies. Furthermore, we discuss the potential challenges to realize the IMSD and provide some promising solutions. K E Y W O R D S deep learning, multi-domain networks, SDN, SFC deployment 1 | INTRODUCTION Recently, a great variety of middleboxes have been designed to efficiently provide various network services. 1 However, the service provisioning method is not flexible and leads to large capital expenditure (CAPEX) and operational expenditure (OPEX) since the middleboxes are heavily dependent on specific hardware devices. Fortunately, the emergence of network function virtualization (NFV) technology provides a promising solution to address the issues. NFV decouples network functions from hardware devices and implements them in software, called virtual network functions (VNFs). 2,3 Unlike the middleboxes, the VNFs can be flexibly deployed in the commercial-off-the-shelf devices. NFV

show abstract

Knowledge-Based Autonomous Service Provisioning in Multi-Domain Elastic Optical Networks

Cited by 52 publications

References 14 publications

Machine Learning for QoT Estimation of Unseen Optical Network States

Machine Learning for QoT Estimation of Unseen Optical Network States

Decentralizing machine-learning-based QoT estimation for sliceable optical networks

The intelligent multi‐domain service function chain deployment: Architecture, challenges and solutions

Contact Info

Product

Resources

About