Skip to main content

Resultados relacionados

Resultados totales (Incluyendo duplicados): 101
Encontrada(s) 2 página(s)

Biblos-e Archivo. Repositorio Institucional de la UAM
Publication Contribución de congreso (ConferenceOutput)

A Latency-Aware Real-Time Video Surveillance Demo: Network Slicing for Improving Public Safety

  • Shariati, B.
  • Dochhan, A.
  • Muqaddas, A.S.
  • Lent, B.
  • Patri, S.
  • Giorgetti, A.
  • Cugini, F.
  • Sgambelluri, A.
  • Nejabati, R.
  • Simeonidou, D.
  • Braun, R.P.
  • Autenrieth, A.
  • Elbers, J.P.
  • Fischer, J.K.
  • Freund, R.
  • Pedreno-Manresa, J. J.
  • Casellas, R.
  • González de Dios, O.
  • Canto, L.L.
  • Moreno, F.J.
  • Pavón, P.
  • Velasco, L.
  • López de Vergara Méndez, Jorge Enrique
  • López Buedo, Sergio
© 2021 IEEE.  Personal use of this material is permitted.  Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works, We report the automated deployment of 5G services across a latency-aware, semidisaggregated, and virtualized metro network. We summarize the key findings in a detailed analysis of end-to-end latency, service setup time, and soft-failure detection time, The research leading to these results has received funding from the EC and BMBF through the METRO-HAUL project (G.A. No. 761727) and OTB-5G+ project (reference No. 16KIS0979K)
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Demonstration of monitoring and data analytics-triggered reconfiguration in partially disaggregated optical networks

  • Gifré Renom, Lluís
  • Boitier, Fabien
  • Delezoide, Camille
  • Ruiz Ramírez, Marc|||0000-0001-6429-6347
  • Buffa, Marta
  • Morea, Annalisa
  • Casellas Regi, Ramon
  • Velasco Esteban, Luis Domingo|||0000-0002-7345-296X
  • Layec, Patricia
©2020 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., We demonstrate a novel agent for optical disaggregated optical networks. When the Monitoring and Data Analytics detects a degradation, it recommends the SDN controller to trigger a network reconfiguration computed by a novel planning tool., The research leading to these results has received funding from the EC through the METRO-HAUL (G.A. nº 761727)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Autonomic content delivery network service

  • Velasco Esteban, Luis Domingo|||0000-0002-7345-296X
  • Gifré Renom, Lluís
  • Ruiz Ramírez, Marc|||0000-0001-6429-6347
© 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., We focus on a use case where a virtualized Content Delivery Network (CDN) service autonomously adapts to the load by requesting the instantiation of new VMs in selected leaf cache nodes, as well as by incrementing the capacity of the network connecting users with the caches. Autonomous decisions are supported by the use of data analytics techniques on monitoring data. Experimental validation supports the feasibility of the proposed autonomic CDN service., This work was partially supported by the EC through the METRO-HAUL project (G.A. nº 761727), from the AEI/FEDER TWINS project (TEC2017-90097-R), and from the Catalan Institution for Research and Advanced Studies (ICREA)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Collective self-learning by exchanging ML models

  • Ruiz Ramírez, Marc|||0000-0001-6429-6347
  • Boitier, Fabien
  • Layec, Patricia
  • Velasco Esteban, Luis Domingo|||0000-0002-7345-296X
Collective self-learning based on Machine Learning (ML) model sharing and combination is proposed to accelerate ML-based algorithm deployment. The considered architecture is presented, together with different alternatives for combining ML models. Performance analysis is carried out on an illustrative use case for autonomic optical transmission., The research leading to these results has received funding from the AEI/FEDER TWINS project (TEC2017-90097-R), from the EC METRO-HAUL project (G.A. nº 761727), and from the Catalan ICREA Institution., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Soft-failure localization and device working parameters estimation in disaggregated scenarios

  • Barzegar, Sima|||0000-0003-1916-7217
  • Virgillito, E.
  • Ruiz Ramírez, Marc|||0000-0001-6429-6347
  • Ferrari, A.
  • Napoli, Antonio
  • Curri, Vittorio
  • Velasco Esteban, Luis Domingo|||0000-0002-7345-296X
A soft-failure localization and key working parameters estimation system is proposed for network diagnosis and maintenance. We show that a double analysis of monitoring data and estimated working parameters greatly anticipates degradations, The research leading to these results has received funding from the EC through the METRO-HAUL project (G.A. nº 761727), from the Spanish MINECO TWINS project (TEC2017-90097-R), and from the Catalan Institution for Research and Advanced Studies (ICREA), Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Quantifying dynamic traffic in elastic optical networks

  • Comellas Colomé, Jaume|||0000-0002-9129-0562
  • Vicario, L.
  • Junyent Giralt, Gabriel|||0000-0003-0059-7700
This work is focused on the characterization of the traffic allocated in Elastic Optical backbone networks when different dynamic traffic profiles are generated. Extensive simulations are done to statistically model the real-time traffic carried by this kind of networks. Well known topologies such as NSFNet and US_Backbone are considered as well as different parameters applied to the traffic generation. Some interesting conclusions are extracted about the properties of the traffic supported by Elastic Optical Networks under dynamic conditions., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Self-learning approaches for real optical networks

  • Ruiz Ramírez, Marc|||0000-0001-6429-6347
  • Boitier, Fabien
  • Layec, Patricia
  • Velasco Esteban, Luis Domingo|||0000-0002-7345-296X
© 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., Self-learning approaches to facilitate the deployment of ML algorithms in real networks are analyzed and their performance evaluated through an illustrative use case. Results show large benefits of collective self-learning with centralized retraining., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Monitoring and data analytics: Analyzing the optical spectrum for soft-failure detection and identification

  • Shariati, Mohammad Behnam
  • Vela, Alba P.
  • Ruiz Ramírez, Marc|||0000-0001-6429-6347
  • Velasco Esteban, Luis Domingo|||0000-0002-7345-296X
Failure detection is essential in optical networks as a result of the huge amount of traffic that optical connections support. Additionally, the cause of failure needs to be identified so failed resources can be excluded from the computation of restoration paths. In the case of soft-failures, their prompt detection, identification, and localization make that recovery can be triggered before excessive errors in optical connections translate into errors on the supported services or even become disrupted. Therefore, Monitoring and Data Analytics (MDA) become of paramount importance in the case of soft-failures. In this paper, we review a MDA architecture that reduces remarkably detection and identification times, while facilitating failure localization. In addition, we rely on Optical Spectrum Analyzers (OSA) deployed in the optical nodes as monitoring devices acquiring the optical spectrum of outgoing links. Analyzing the optical spectrum of optical connections, specific soft-failures that affect the shape of the spectrum can be detected. A workflow consisting of machine learning algorithms, designed to be integrated in the aforementioned MDA architecture, will be studied to analyze the optical spectrum of a given optical connection acquired in a node and to determine whether a filter failure is affecting it, and in such case, what is the type of filter failure and its magnitude. Exhaustive results are presented allowing to evaluate the proposed method. © 2018 IFIP., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Bringing data analytics to the network nodes for efficient traffic anomalies detection

  • Pérez Vela, Alba|||0000-0001-6301-0749
  • Ruiz Ramírez, Marc|||0000-0001-6429-6347
  • Velasco Esteban, Luis Domingo|||0000-0002-7345-296X
Traffic anomalies can create network congestion, so its prompt and accurate detection would allow network operators to make decisions to guarantee the network performance avoiding services to experience any perturbation. In this paper, we focus on origin-destination (OD) traffic anomalies; to efficiently detect those, we study two different anomaly detection methods based on data analytics and combine them with three monitoring strategies. In view of the short monitoring period needed to reduce anomaly detection, which entails large amount of monitoring data to be collected and analyzed in a centralized repository, we propose bringing data analytics to the network nodes to efficiently detect traffic anomalies, while keeping traffic estimation centralized. Exhaustive simulation results on a realistic network scenario show that the monitoring period should be as low as possible (e.g., 1 min) to keep anomaly detection times low, which clearly motivates to place traffic anomaly detection function in the network nodes., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Autonomic transmission through pre-FEC BER degradation prediction based on SOP monitoring

  • Shariati, Mohammad Behnam
  • Boitier, F.
  • Ruiz Ramírez, Marc|||0000-0001-6429-6347
  • Layec, Patricia
  • Velasco Esteban, Luis Domingo|||0000-0002-7345-296X
An Autonomic Transmission Agent based on machine-learning is proposed for excessive bit-error-rate prediction resulting from real-time analysis of state-of-polarization (SOP). The accuracy and speed of the agent enables its use for local and remote transceiver reconfiguration., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Applying data visualization for failure localization

  • Vela, Alba P.
  • Ruiz Ramírez, Marc|||0000-0001-6429-6347
  • Velasco Esteban, Luis Domingo|||0000-0002-7345-296X
Data visualization is applied to BER measures in a bigdata repository. Bubble charts are produced to identify lightpaths with increasing BER and spectrum color maps are then used to identify the most likely degraded link, Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Knowledge management in optical networks

  • Velasco Esteban, Luis Domingo|||0000-0002-7345-296X
  • Tabatabaeimehr, Fatemehsadat|||0000-0003-4498-4219
  • Ruiz Ramírez, Marc|||0000-0001-6429-6347
©2020 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., Autonomous network operation realized by means of control loops, where prediction from machine learning (ML) models is used as input to proactively reconfigure individual optical devices or the whole optical network, has been recently proposed to minimize human intervention. A general issue in this approach is the limited accuracy of ML models due to the lack of real data for training the models. Although the training dataset can be complemented with data from lab experiments and simulation, it is probable that once in operation, events not considered during the training phase appear thus leading into model inaccuracies. A feasible solution is to implement self-learning approaches, where model inaccuracies are used to re-train the models in the field and to spread such data for training models being used for devices of the same type in other nodes in the network. In this paper, we develop the concept of collective self-learning aiming at improving models error convergence time, as well as at minimizing the amount of data being shared and stored. To this end, we propose a knowledge management (KM) process and an architecture to support it., The research leading to these results has received funding from the European Commission through the METROHAUL project (G.A. nº 761727), from the Spanish MINECO TWINS project (TEC2017-90097-R), and from the Catalan Institution for Research and Advanced Studies (ICREA)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

CASTOR: a monitoring and data analytics architecture to support autonomic domain and slice networking

  • Vela, A.P.
  • Gifre Renom, Lluís
  • González de Dios, Oscar
  • Ruiz Ramírez, Marc|||0000-0001-6429-6347
  • Velasco Esteban, Luis Domingo|||0000-0002-7345-296X
Network slices combine resource virtualization with the isolation level required by future 5G applications. In addition, the use of Monitoring and Data Analytics (MDA) helps to maintain the required network performance, while reducing total cost of ownership. In this paper, we present CASTOR, an architecture to enable autonomic domain and slice networking. MDA agents use data analytics to make local decisions close to network devices, whereas MDA controllers collate and export metered data transparently to customer controllers, all of them leveraging customizable and isolated data analytics processes. Discovered knowledge can be applied for both proactive and reactive network slice reconfiguration, triggered either by service providers or customers, thanks to the interaction with state-of-the-art software-defined networking controllers and planning tools. © 2018 IEEE., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Interfaces for monitoring and data analytics systems

  • Gifré Renom, Lluís
  • Ruiz Ramírez, Marc|||0000-0001-6429-6347
  • Velasco Esteban, Luis Domingo|||0000-0002-7345-296X
© 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., Software defined networks (SDN) represents one of the most relevant innovations in recent years for the telecom industry. Major operators are planning to progressively migrate their transport networks, including optical, to such paradigm to take advantage of the programmability of connectivity services. In SDN, resources are abstracted at the networking level and exposed via a north-bound interface to management systems. On the other hand, Monitoring and Data Analytics (MDA) systems can collect monitoring data and make them accessible to external systems, such as billing platforms, to accurately charge users for the services they consume. In addition, in dynamic scenarios, where resources could be created, modified and removed on demand, MDA systems can also facilitate autonomic networking, and thus coordination between SDN and MDA is strictly required. In this paper, we present two monitoring interfaces for MDA systems: the Monitoring (M)-Control, Orchestration, and Management (COM) interface, an east-west interface that enables MDA systems to gather and synchronize operational and telemetry information from multiple COM modules, e.g., an SDN controller or a cloud resource manager, and issue recommendations to the COM modules as a result of data analysis; and the IO4 interface, a MDA north-bound interface that facilitates exporting monitored data to external systems. Functionality of each interface is described from the conceptual viewpoint and relevant operational workflows are proposed., This work was partially supported by the EC through the METRO-HAUL project (G.A. nº 761727), from the AEI/FEDER TWINS project (TEC2017-90097-R), and from the Catalan ICREA Institution., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Allocating multi-rate requests to single-rate lightpaths in Flexgrid optical networks

  • Comellas Colomé, Jaume|||0000-0002-9129-0562
  • Junyent Giralt, Gabriel|||0000-0003-0059-7700
Some ideas about the effect of appropriately selecting the size of the connections conveyed by multi-rate flexgrid optical networks are discussed. The main focus is on illustrating the relation between spectrum consumption and effective carried traffic. It is demonstrated that by introducing some order in the spectrum use, so limiting the network flexibility, its global performance is notably improved. The main contribution consists in demonstrating that using single sized lightpaths, the network throughput at any given blocking probability value increases compared to the multi-rate case. This conclusion is supported by extensive simulations under dynamic traffic conditions, considering different network topologies and traffic profiles., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Learning from the optical spectrum: soft-failure identification and localization

  • Velasco Esteban, Luis Domingo|||0000-0002-7345-296X
  • Shariati, Mohammad Behnam
  • Vela, Alba P.
  • Comellas Colomé, Jaume|||0000-0002-9129-0562
  • Ruiz Calvo, Manuel
The availability of coarse-resolution cost-effective Optical Spectrum Analyzers (OSA) allows its widespread deployment in operators' networks. In this paper, several machine learning approaches for failure identification and localization that take advantage of OSAs are presented, Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Automated end to end Carrier Ethernet provisioning over a disaggregated WDM metro network with a hierarchical SDN control and monitoring platform

  • Morro, Roberto
  • Lucrezia, Francesco
  • Gomes, P.
  • Casellas Regi, Ramón
  • Giorgetti, Alessio
  • Velasco Esteban, Luis Domingo|||0000-0002-7345-296X
  • Riccardi, Emilio
  • Chiadò Piat, Anna
  • Percelsi, Alessandro
  • Pedro, João|||0000-0003-4471-7401
  • Gifre Renom, Lluís
  • Sgambelluri, Andrea
  • Risso, Fulvio
  • Marchetto, Guido
This demo shows how a hierarchical control plane of ONOS SDN controllers orchestrates the dynamic provisioning of end-to-end Carrier Ethernet circuits on a composite network, programming the whole data path from the CPE to the core optical equipment., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Enabling data analytics and machine learning for 5G services within disaggregated multi-layer transport networks

  • Casellas Regi, Ramón
  • Martínez Rivera, Ricardo Victor
  • Velasco Esteban, Luis Domingo|||0000-0002-7345-296X
  • Vilalta, Ricard
  • Pavón Mariño, Pablo
  • King, Daniel
  • Muñoz González, Raül
Recent advances, related to the concepts of Artificial Intelligence (AI) and Machine Learning (ML) and with applications across multiple technology domains, have gathered significant attention due, in particular, to the overall performance improvement of such automated systems when compared to methods relying on human operation. Consequently, using AI/ML for managing, operating and optimizing transport networks is increasingly seen as a potential opportunity targeting, notably, large and complex environments.Such AI-assisted automated network operation is expected to facilitate innovation in multiple aspects related to the control and management of future optical networks and is a promising milestone in the evolution towards autonomous networks, where networks self-adjust parameters such as transceiver configuration.To accomplish this goal, current network control, management and orchestration systems need to enable the application of AI/ML techniques. It is arguable that Software-Defined Networking (SDN) principles, favouring centralized control deployments, featured application programming interfaces and the development of a related application ecosystem are well positioned to facilitate the progressive introduction of such techniques, starting, notably, in allowing efficient and massive monitoring and data collection.In this paper, we present the control, orchestration and management architecture designed to allow the automatic deployment of 5G services (such as ETSI NFV network services) across metropolitan networks, conceived to interface 5G access networks with elastic core optical networks at multi Tb/s. This network segment, referred to as Metro-haul, is composed of infrastructure nodes that encompass networking, storage and processing resources, which are in turn interconnected by open and disaggregated optical networks. In particular, we detail subsystems like the Monitoring and Data Analytics or the in-operation planning backend that extend current SDN based network control to account for new use cases., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Experimental assessment of big data-backed video distribution in the telecom cloud

  • Gifre Renom, Lluís
  • Ruiz Ramírez, Marc|||0000-0001-6429-6347
  • Velasco Esteban, Luis Domingo|||0000-0002-7345-296X
The telecom infrastructure is undergoing a huge transformation since operators are deploying their own cloud infrastructure to provide cloud services and enabling Network Functions Virtualization (NFV). The resulting infrastructure is referred to as the telecom cloud. NFV decouples network functions from proprietary hardware appliances, so they can be implemented in software and deployed on virtual machines (VM) running on commercial off-the-shelf computing hardware. One of the network functions currently being considered by many operators is Content Delivery Network (CDN) for live-TV and Video on Demand (VoD) distribution. In fact, a significant advance in video delivery is the standardized MPEG Dynamic Adaptive Streaming over HTTP (MPEG-DASH) technique that enables media content delivering over the Internet using standard HTTP web server infrastructure. Besides, the CDN reconfiguration, motivated by an increase in the number of users demanding content, entails a Virtual Network Topology (VNT) reconfiguration to cope with the demanded connection capacities. A Big Data-based CDN manager, fulfilling the ETSI NFV guidelines, is proposed to adapt the virtualized CDN function to current and future demand. Its feasibility to control virtualized components while collecting monitoring data is experimentally assessed in a real environment., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Optical signal tracking for robust PAM4 deployment in filterless metro network scenarios

  • Shariati, Mohammad Behnam
  • Fresi, Francesco
  • Ruiz Ramírez, Marc|||0000-0001-6429-6347
  • Cugini, Filippo
  • Velasco Esteban, Luis Domingo|||0000-0002-7345-296X
© 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., Highly accurate and reliable optical signal tracking is proposed that estimates sub-GHz laser drift failures by analyzing spectra acquired by cost-effective coarse-granular OSAs. Its application on PAM4 systems in filterless metro networks brings added robustness., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Monitoring and data analytics-triggered reconfiguration in partially disaggregated optical networks

  • Gifré Renom, Lluís
  • Boitier, Fabien
  • Delezoide, Camille
  • Ruiz Ramírez, Marc|||0000-0001-6429-6347
  • Buffa, Marta
  • Morea, Annalisa
  • Casellas Regi, Ramon
  • Velasco Esteban, Luis Domingo|||0000-0002-7345-296X
  • Layec, Patricia
©2020 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., We present ADONIS, which stands for Aggregator/Disaggregator for Optical Network equIpmentS, a novel open device agent able to construct logical network devices from (dis)aggregation of physical components in order to expose meaningful network devices to the SDN controller. We experimentally assess it by means of a control closed-loop involving ADONIS, a Software Defined Network controller, a Monitoring and Data Analytics system, and a novel reconfiguration tool, SMART-A., The research leading to these results has received funding from the EC through the METRO-HAUL (G.A. nº 761727)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Toward multilayer disaggregated node telemetry and local decision making

  • Velasco Esteban, Luis Domingo|||0000-0002-7345-296X
  • Gifre Renom, Lluís
  • Izquierdo Zaragoza, José Luis|||0000-0003-4013-9712
A generic node agent supporting disaggregated node telemetry is presented. Data collection close to devices enable making local decisions, leveraging SDN controllers for network-wide operations. The agent is demonstrated in a BER-triggered transponder reconfiguration scenario., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

On the benefits of elastic spectrum management in multi-hour filterless metro networks

  • Pedreno Manresa, J.J.
  • Izquierdo Zaragoza, José Luis|||0000-0003-4013-9712
  • Cugini, Filippo
  • Pavón Mariño, Pablo
The dawn of 5G is pushing operators to deploy high-capacity, agile networks capable of adapting to time-varying traffic patterns, especially into metro sections. ROADMs are key enablers for agility in the optical layer, however the benefits of this agility do not always compensate for increased costs. As such, filterless optical networks are emerging as a cost-effective and reliable solution compared to active photonics, thanks to a winning combination of coherent transponders and passive splitters/couplers. However, spectrum allocation management policies are of paramount importance to maximize the overall network throughput. In this paper, we focus on a filterless metro network where the hourly variation of the demands traffic is known, coming from historic data estimations. Then, we observe how the knowledge of the traffic profiles can be exploited. To assess this, we evaluate the performance, in terms of throughput, of three different spectrum management approaches: (i) fixed, where lightpaths remain static along time once allocated; (ii) semi-elastic, where lightpath-bandwidth vary according to current traffic requirements, but central frequency remains fixed; and (iii) hitless full-elastic, where any lightpath parameter may be reconfigured without disrupting the traffic. Besides, we consider two transponder types equipped with (i) shared or (ii) independent tunable lasers for transmission and reception, which affects to spectrum allocation of bidirectional connections. According to our results, the semi-elastic approach clearly outperforms the fixed approach (23-33% more throughput) with a reduced gap to the hitless full-elastic case (10-24% less throughput), especially considering that the latter is not commercially available yet. Interestingly, using dual-laser transponders only yields a 10% gain with respect to single-laser transponders for the semi-elastic scenario, and thus may not justify the extra hardware., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Soft failure localization in elastic optical networks

  • Vela, A.P.
  • Shariati, Mohammad Behnam
  • Ruiz Ramírez, Marc|||0000-0001-6429-6347
  • Comellas Colomé, Jaume|||0000-0002-9129-0562
  • Velasco Esteban, Luis Domingo|||0000-0002-7345-296X
Soft failure localization to early detect service level agreement violations is of paramount importance in elastic optical networks (EONs), while it allows anticipating possible hard failure events. Nowadays, effective and automated solutions for soft failure localization during lightpaths' commissioning testing and operation are still missing. In this paper, we focus on presenting soft failure localization algorithms based on two different active monitoring techniques. First, the Testing optIcal Switching at connection SetUp timE (TISSUE) algorithm is proposed to localize soft failures during commissioning testing phase by elaborating the estimated bit-error rate (BER) values provided by low-cost optical testing channel (OTC) modules. Second, the FailurE causE Localization for optIcal NetworkinG (FEELING) algorithm is proposed to localize failures during lightpath operation using cost-effective optical spectrum analyzers (OSAs) widely deployed in network nodes. Results are presented to validate both algorithms in the event of several soft failures affecting lasers and filters. © 2018 IEEE., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Data analytics based origin-destination core traffic modelling

  • Morales Alcaide, Fernando|||0000-0001-7337-1775
  • Ruiz Ramírez, Marc|||0000-0001-6429-6347
  • Velasco Esteban, Luis Domingo|||0000-0002-7345-296X
Traffic monitoring is an essential task for network operators since it allows evaluating network performance. Monitoring data from origin-destination (OD) traffic in core virtual network topologies can be collected from packet nodes and stored in a repository for further analysis, e.g., to detect anomalies or to create predicted traffic matrices for the near future. In this paper we propose a set of modules to support data analytics-based algorithms along with a machine learning procedure based on artificial neural networks (ANN) that provides robust and adaptive traffic models., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Periodic defragmentation in elastic optical networks

  • Comellas Colomé, Jaume|||0000-0002-9129-0562
  • Vicario, L.
  • Junyent Giralt, Gabriel|||0000-0003-0059-7700
The main weakness of Elastic Optical Networks (EON) stems from spectrum fragmentation. An analysis of periodic defragmentation in EONs under dynamic traffic conditions is carried out. The effects of different defragmentation parameters on the EON performance are evaluated, and appropriate values, guaranteeing suitable network performance while keeping the network control complexity at a reasonable point, are obtained through simulations. Different network topologies as well as traffic conditions are simulated to assess the validity of the obtained results., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Generation and analysis of service-based traffic flows

  • Bernal Escribano, Álvaro
  • Ruiz Ramírez, Marc|||0000-0001-6429-6347
  • Velasco Esteban, Luis Domingo|||0000-0002-7345-296X
© 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., The rapid availability of new services makes that network operators cannot exhaustively test their impact on the network or anticipate any capacity exhaustion. This situation will be worse with the imminent introduction of the 5G technology and the kind of totally new services that it will support. In this paper, we present CURSA-SQ, a methodology to analyze the network behavior when the specific traffic that would be generated by groups of service consumers is injected. CURSA-SQ includes input traffic flow modelling with second and sub-second granularity based on specific service and consumer behaviors. The methodology allows to accurately study traffic flows at the input and outputs of complex scenarios with multiples queues systems, as well as other metrics such as delays., This work was partially supported by the EC through the METRO-HAUL project (G.A. nº 761727), from the AEI/FEDER TWINS project (TEC2017-90097-R), and from the Catalan ICREA Institution., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Experimental assessment of SDN-enabled reconfigurable OPSquare data center networks with QoS guarantees

  • Xue, Xuwei
  • Wang, Fulong
  • Agraz Bujan, Fernando|||0000-0001-8958-0088
  • Pagès Cruz, Albert|||0000-0002-6661-3630
  • Pan, B
  • Yan, F
  • Spadaro, Salvatore|||0000-0002-4100-1726
  • Calabretta, Nicola
An SDN-controlled DCN enabling QoS-driven network-slice reconfiguration and packet priority updating is experimentally assessed. Network-slice can be reconfigured within 150 ms to decrease packet loss. The measured ToR-to-ToR latency of high-priority packets is 260.55 ns., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Flexible fog computing and telecom architecture for 5G networks

  • Velasco Esteban, Luis Domingo|||0000-0002-7345-296X
  • Ruiz Ramírez, Marc|||0000-0001-6429-6347
We review a novel, secure, highly distributed and ultra-dense fog computing infrastructure, which can be allocated at the extreme edge of a wired/wireless network for a Telecom Operator to provide multiple unified, cost-effective and new 5G services, such as Network Function Virtualization (NFV), Mobile Edge Computing (MEC), and services for third parties (e.g., smart cities, vertical industries or Internet of Things (IoT)). The distributed and programmable fog technologies are expected to strengthen the position of the Mobile Network and cloud markets; key benefits are the dynamic deployment of new distributed low-latency services. The architecture consists of three main building blocks: a) a scalable node, that is seamlessly integrated in the Telecom infrastructure; b) a controller, focused on service assurance, that is integrated in the management and orchestration architecture of the Telecom operator; and c) services running on top of the Telecom infrastructure., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Visualization tools for enhancing failure localization [Invited]

  • Pérez Vela, Alba|||0000-0001-6301-0749
  • Ruiz Ramírez, Marc|||0000-0001-6429-6347
  • Velasco Esteban, Luis Domingo|||0000-0002-7345-296X
© 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., The use of data visualization techniques is explored to guide human operators in failure localization tasks. In this paper, we consider the case where a subset of lightpaths’ experience gradual BER degradation. We aim at using visualization techniques to find out the most likely resource that is responsible for such degradation, as well as the start of the degradation. Note that the challenge is to analyze an amount of data that might be huge (e.g., 1 sample every 15 min for the last, let us say, 4 weeks, for every lightpath in the network) and display charts in seconds, while revealing meaningful information for human operators. The proposed task-oriented visualization tackles such challenge by using charts that first help finding the subset of affected lightpaths thus, reducing the number of lightpaths to be analyzed. In a second step, different charts are used to reduce the element in failure as much as possible. Finally, the few suspicious elements can be analyzed in detail using more traditional timeline graphs., The research leading to these results has received funding from the Spanish MINECO TWINS project (TEC2017-90097-R), from the EC through the METRO-HAUL project (G.A. no 761727), and from the Catalan Institution for Research and Advanced Studies (ICREA)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Experimental demonstration of active and passive optical networks telemetry

  • Gifre Renom, Lluís
  • Izquierdo Zaragoza, José Luis|||0000-0003-4013-9712
  • Shariati, Mohammad Behnam
  • Velasco Esteban, Luis Domingo|||0000-0002-7345-296X
A distributed architecture enabling active and passive optical network telemetry is presented; a YANG data model is used for remotely configuring monitoring devices for telemetry purposes. Experimental demonstration is carried out for failure localization., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Distributed and centralized options for self-learning

  • Tabatabaeimehr, Fatemehsadat|||0000-0003-4498-4219
  • Ruiz Ramírez, Marc|||0000-0001-6429-6347
  • Velasco Esteban, Luis Domingo|||0000-0002-7345-296X
© 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., In general, the availability of enough real data from real fog computing scenarios to produce accurate Machine Learning (ML) models is rarely ensured since new equipment, techniques, etc., are continuously being deployed in the field. Although an option is to generate data from simulation and lab experiments, such data could not cover the whole features space, which would translate into ML models inaccuracies. In this paper, we propose a self-learning approach to facilitate ML deployment in real scenarios. A dataset for ML training can be initially populated based on the results from simulation and lab experiments and once ML models are generated, ML re-training can be performed after inaccuracies are detected to improve their precision. Illustrative numerical results show the benefits from the proposed self-learning approach for two general use cases of regression and classification., This work was partially supported by the EC through the METRO-HAUL project (G.A. nº 761727), from the AEI/FEDER TWINS project (TEC2017-90097-R), and from the Catalan ICREA Institution., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

A latency-aware real-time video surveillance demo: network slicing for improving public safety

  • Shariati, Mohammad Behnam
  • Pedreno Manresa, José Juan
  • Dochhan, Annika
  • Siddique Muqaddas, Abubakar
  • Casellas Regi, Ramon
  • González de Dios, Oscar
  • Luque Canto, L.
  • Lent, Bodo
  • López de Vergara Méndez, Jorge Enrique
  • Velasco Esteban, Luis Domingo|||0000-0002-7345-296X
We report the automated deployment of 5G services across a latency-aware, semi- disaggregated, and virtualized metro network. We summarize the key findings in a detailed analysis of end-to-end latency, service setup time, and soft-failure detection time., The research leading to these results has received funding from the EC and BMBF through the METRO-HAUL project (G.A. No. 761727) and OTB-5G+ project (reference No. 16KIS0979K)., Peer Reviewed, Article signat per 24 autors/es: B. Shariati, Fraunhofer HHI, Germany / J. J. Pedreno-Manresa, ADVA, Germany / A. Dochhan, ADVA, Germany / A. S. Muqaddas, UOB, UK / R. Casellas, CTTC/CERCA, Spain / O. González de Dios, Telefónica, Spain / L. L. Canto, Telefónica, Spain / B. Lent, Qognify GmbH, Germany / J. E. López de Vergara, Naudit HPCN, Spain / S. López-Buedo, Naudit HPCN, Spain / F. J. Moreno, UPCT, Spain / P. Pavón, UPCT, Spain / L. Velasco, UPC, Spain / S. Patri, ADVA, Germany / A. Giorgetti, CNIT, Italy / IEIIT-CNR, F. Cugini CNIT, Italy / A. Sgambelluri, CNIT, Italy / R. Nejabati, UOB, UK / D. Simeonidou, UOB, UK / R.-P. Braun, Deutsche Telekom, Germany / A. Autenrieth, ADVA, Germany / J.-P. Elbers, ADVA, Germany / J. K. Fischer, Fraunhofer HHI, Germany / R. Freund, Fraunhofer HHI, Germany
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Adaptation of the residual signal for filter failure detection in scenarios with multiple filter types

  • Barzegar, Sima|||0000-0003-1916-7217
  • Ruiz Ramírez, Marc|||0000-0001-6429-6347
  • Velasco Esteban, Luis Domingo|||0000-0002-7345-296X
© 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., To monitor the optical spectra of outgoing links, Optical Spectrum Analyzers (OSA) can deployed in the optical nodes acquiring the optical spectra of outgoing links and analyzing optical signals to detect those soft-failures that affect the shape of the spectrum, like Filter Shift and Filter Tightening. In this paper, we rely on the residual-based approach spectra analysis that uses a set of residual signals calculated by subtracting the measured signal by the OSA from an expected version of the signal. In this paper, we assume a network scenario, where nodes with optical filters of different characteristics (e.g., 2nd, 3rd, and 4th order Gaussian) are installed. In this scenario, the possibility of generating complete data set for machine learning training purposes that include every combination of filter and failure is difficult to achieve. Instead, we propose an adaptation mechanism that transform the expected effects of every filter type thus, emulating single filter-type networks. Exhaustive experiments are carried out to show the performance of the proposed adaptation under filter failures., This work was partially supported by the EC through the METRO-HAUL project (G.A. nº 761727), from the AEI/FEDER TWINS project (TEC2017-90097-R), and from the Catalan ICREA Institution., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Impact of traffic delay tolerance on elastic optical networks performance

  • Comellas Colomé, Jaume|||0000-0002-9129-0562
  • Nourmohammadi, Farzaneh
  • Junyent Giralt, Gabriel|||0000-0003-0059-7700
© 2020 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., Elastic Optical Networks (EON) are considered as a valuable solution to enhance optical networks efficiency due to their superior network spectrum use. By breaking the traditional fixed grid of WDM networks, the spectrum assigned to connection demands is adapted to their requirements. On the other hand, connection parameters such as baud-rate or modulation format are dynamically adjusted taking into account the current network conditions. Therefore, EONs fit perfectly with the heterogeneous characteristics of current and future network traffic loads. Nevertheless, when EONs are in operation, they present some weaknesses mainly derived from the spectrum fragmentation generated by the dynamic establishment and torn-down of these diverse connections. The randomness of connections generation results in the blocking of some connections even when the total load supported by the network is not so high. In this work, we analyse how traffic tolerance to delay impacts on the EON performance, and present simulation obtained results when connection demands have no strict delay constraints compared with demands which need to be real-time served., This work was partially supported by the EC through the METRO-HAUL project (G.A. 761727), and from the Spanish MINECO TWINS project (TEC2017-90097-R)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Spectrally and spatially flexible optical networks: recent developments and findings

  • Shariati, Mohammad Behnam
  • Klonidis, Dimitrios
  • Marom, Dan M.
  • Comellas Colomé, Jaume|||0000-0002-9129-0562
  • Velasco Esteban, Luis Domingo|||0000-0002-7345-296X
  • Tomkos, Ioannis
Space division multiplexing (SDM) has been proposed as the ultimate solution to address the capacity crunch of optical transport networks. The efficient utilization of SDM requires some forms of spatial integration which is expected to bring huge cost savings for the deployment of SDM-based optical networks. Spatially integrated components with different characteristics have been largely explored and demonstrated for the realization of SDM networks, including; transmission media, transceivers with sophisticated digital signal processing (DSP) units, amplifiers, and ultimately optical switching nodes which are the key elements for the realization of transparent optical networks exploiting SDM technologies. As a consequence, in contrast to the currently deployed optical networks based on standard single mode fibers (SSMF), the next generation of optical networks exploiting SDM technologies can be realized utilizing various kinds of transmission media and the other corresponding elements. However, due in part to the physical properties of different components, their complexity, and the technology limitations not all of them meet the economic feasibility and a desired level of network-wide performance. Therefore, careful analysis should be done revealing the most appropriate and cost-effective solutions. In this work, we review the recent developments and findings which pave the way for a simplified and efficient implementation of the spectrally and spatially flexible optical networks., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

CASTOR: An architecture to bring cognition to transport networks

  • Velasco Esteban, Luis Domingo|||0000-0002-7345-296X
  • Gifre Renom, Lluís
  • Izquierdo Zaragoza, José Luis|||0000-0003-4013-9712
  • Julián Moreno, Guillermo
  • López de Vergara Méndez, Jorge E.
CASTOR architecture to enable cognitive networking is demonstrated. Extended nodes make local decisions, whilst a centralized system beside the network controller makes network-wide decisions. Interaction with ONOS, Net2Plan, and passive monitoring devices is exhibited., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Fault management based on machine learning

  • Velasco Esteban, Luis Domingo|||0000-0002-7345-296X
  • Rafique, Danish
© 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., Machine Learning (ML) brings many benefits for network operation. In this paper, basic ML concepts and its integration into existing network control and management planes are reviewed. Case studies covering fault management are illustrated., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Performance evaluation of the VNT reconfiguration algorithm based on traffic prediction

  • Morales Alcaide, Fernando|||0000-0001-7337-1775
  • Festa, Paola
  • Ruiz Ramírez, Marc|||0000-0001-6429-6347
  • Velasco Esteban, Luis Domingo|||0000-0002-7345-296X
In a previous work, the Virtual Network Topology Reconfiguration problem based on Traffic Prediction (VENTURE) was proposed as a means of efficiently adapting core virtual network topology (VNT) to the near-future traffic. Although the benefits obtained by the VENTURE algorithm compared to using a purely reactive VNT reconfiguration approach seems to be clear, margin for improvement still remain and alternative solving methods for the VENTURE problem need to be considered. In this paper, the original VENTURE algorithm is compared against two state-of-the-art metaheuristics for combinatorial network optimization. The two metaheuristics are first presented and then adapted for the VENTURE problem. Finally, the performance of the VENTURE algorithm and the two proposed metaheuristics is numerically evaluated using an exact solving method as reference. © 2018 IEEE.
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Combining a machine learning and optimization for early pre-FEC BER degradation to meet committed QoS

  • Pérez Vela, Alba|||0000-0001-6301-0749
  • Ruiz Ramírez, Marc|||0000-0001-6429-6347
  • Cugini, Filippo
  • Velasco Esteban, Luis Domingo|||0000-0002-7345-296X
Monitoring the physical layer is key to detect bit error rate (BER) degradation caused by failures and to identify the cause of the failure and localize the failed elements. Once the failure has been detected, actions can be taken to reduce as much as possible its impact on the network. Commercially available optical equipment are able to correct degraded optical signals by means of Forward Error Correction (FEC) algorithms. A value of pre-FEC BER over a pre-defined threshold would imply a non-error-free post-FEC transmission and, as a result, communication would be disrupted. Therefore, a prompt detection of lightpaths with excessive pre-FEC BER can help to greatly reduce such SLA violations, in particular when supporting vlinks. As a result of the above, it would be desirable to anticipate such degradations and apply re-optimization to re-route those affected demands according to their SLAs in order to reduce the affected traffic after the degradation is detected. Designing algorithms capable of promptly detect distinct BER anomaly patterns would be desirable. The objective would be to anticipate intolerable BER values as much as possible aiming at leaving enough time to plan a re-routing procedure during off-peak hours. In this paper, we propose an effective machine learning-based algorithm to localize and identify the most probable cause of failure impacting a given service, as well as a re-optimization algorithm to re-route affected demands, targeting at reducing SLA violation. Results show that the proposed detection and re-routing algorithms noticeably reduce bandwidth and number of demands affected., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Out-of-field generic ML training with in-field specific adaptation to facilitate ML deployments

  • Shariati, Mohammad Behnam
  • Ruiz Ramírez, Marc|||0000-0001-6429-6347
  • Velasco Esteban, Luis Domingo|||0000-0002-7345-296X
© 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., A two-phase strategy to facilitate ML algorithm deployment in real networks is demonstrated: out-of-field training uses data from simulation and testbed experiments with generic equipment whereas in-field adaptation is applied to support heterogeneous equipment., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Monitoring and data analytics for optical networking

  • Velasco Esteban, Luis Domingo|||0000-0002-7345-296X
  • Ruiz Ramírez, Marc|||0000-0001-6429-6347
© 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., Increased levels of automation will be necessary in view of more stringent performance requirements that next generation optical transport networks need to support in a near term, not only for high capacity, but, even more importantly, for dynamicity, latency, and availability. All these aspects will become more relevant with the growing complexity of modern networks. Network automation targets resource re-optimization to rapidly adapt the network to the expected conditions, quick degradation detection to improve the quality of the connections, as well as failure detection and identification to facilitate maintenance. Network automation requires and implies the collection of data for performance monitoring, being then elaborated by data analytics algorithms to produce meaningful inputs for the network controller, which will finally program the underlying devices. In this paper, we analyze alternative architectures for monitoring and data analytics (MDA) and illustrative control loops are presented aiming at validating the usefulness of MDA to automate optical networks operation., The research leading to these results has received funding from the EC through the METRO-HAUL project (G.A. nº 761727), from the Spanish MINECO TWINS project (TEC2017-90097-R), and from the Catalan Institution for Research and Advanced Studies (ICREA)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Using dual-path allocation for partial traffic protection in elastic optical networks

  • Junyent Giralt, Gabriel|||0000-0003-0059-7700
  • Comellas Colomé, Jaume|||0000-0002-9129-0562
Multi-path allocation has been proposed as a simple way to achieve traffic protection in different networking domains. In this work dual-path allocation is proposed for Elastic Optical Networks (EON) assuming dynamic offered traffic, and its performance is quantified by simulation. Indeed, dual-path allocation is compared with partial dedicated path protection assuming different scenarios. In addition to a qualitative comparison where advantages and drawbacks of both schemes are analysed, quantitative results for different backbone network scenarios are presented., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Adapting the virtual network topology to near future traffic

  • Morales Alcaide, Fernando|||0000-0001-7337-1775
  • Festa, Paola
  • Ruiz Ramírez, Marc|||0000-0001-6429-6347
  • Velasco Esteban, Luis Domingo|||0000-0002-7345-296X
The introduction of new services requiring large and dynamic bitrate connectivity can cause changes in the direction of the traffic in metro and even core network segments along the day. This leads to large overprovisioning in statically managed virtual network topologies (VNT), designed to cope with the traffic forecast. To reduce expenses while ensuring the required grade of service, in this paper we propose the VNT reconfiguration approach based on current and near-future traffic matrices (VENTURE) to regularly adapt the topology to both, the current and future traffic volume and direction. The problem is formally stated and a heuristic algorithm is proposed to solve it., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Smart filterless optical networks based on optical spectrum analysis

  • Ruiz Ramírez, Marc|||0000-0001-6429-6347
  • Sgambelluri, Andrea
  • Cugini, Filippo
  • Velasco Esteban, Luis Domingo|||0000-0002-7345-296X
© 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., Dynamic network operations can produce power fluctuations of the established connections in filterless optical networks. In addition, the gridless nature of filterless networks make that some (un)intentional effects such as transponders laser drift might disrupt the proper operation of lightpaths. To overcome these issues, we present a monitoring system exploiting data analytics and cost-effective optical spectrum analyzers to achieve smart filterless network operation. Experimental measurements are used to validate the proposed data analytics-based approaches, as well as to find the optimal resolution to achieve maximum performance with minimum cost., This work was partially supported by the EC through the METRO-HAUL project (G.A. nº 761727), from the AEI/FEDER TWINS project (TEC2017-90097-R), and from the Catalan ICREA Institution., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Autonomous operations in optical networks

  • Pérez Vela, Alba|||0000-0001-6301-0749
  • Ruiz Ramírez, Marc|||0000-0001-6429-6347
  • Velasco Esteban, Luis Domingo|||0000-0002-7345-296X
© 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., Machine Learning (ML) has already proven its benefits for network operation, being a sub-domain of artificial intelligence, it is highly suitable for complex system representation. In this paper, basic ML concepts are reviewed, as well as its integration into existing network control and management planes. Then, a use case focused on soft-failure detection is presented in detail covering optical spectrum analysis and ML algorithms; the technique relies on the widespread deployment of cost-effective optical spectrum analyzer (OSA). Finally, the retrieved optical parameters are analyzed using ML algorithms giving rise to illustrative results., The research leading to these results has received funding from the Spanish MINECO TWINS project (TEC2017-90097-R), from the EC through the METRO-HAUL project (G.A. no 761727), and from the Catalan Institution for Research and Advanced Studies (ICREA)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Autonomic NFV network services on top of disaggregated optical metro networks

  • Esmenats, P. R.
  • Casellas Regi, Ramon
  • Gifre Renom, Lluís
  • Pérez Vela, Alba|||0000-0001-6301-0749
  • Ruiz Ramírez, Marc|||0000-0001-6429-6347
  • Martínez, Ricardo
  • Velasco Esteban, Luis Domingo|||0000-0002-7345-296X
Control and orchestration of metro network to support dynamic provisioning and reconfiguration of services based on Virtual Network Functions are demonstrated through a CDN example. Interaction among CASTOR MDA, OSM and ONOS is exhibited., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Real-time spectrum surveillance in filterless optical networks

  • Shariati, Mohammad Behnam
  • Ruiz Ramírez, Marc|||0000-0001-6429-6347
  • Sgambelluri, Andrea
  • Cugini, Filippo
  • Velasco Esteban, Luis Domingo|||0000-0002-7345-296X
A monitoring system exploiting data analytics and cost-effective optical spectrum analyzers with under 1.2GHz resolution is proposed. Its performance is demonstrated in a filterless network experimental test-bed. The system enables prompt action before lightpath disruption. © 2018 OSA., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Dynamic virtual network connectivity for C-RAN backhauling

  • Asensio Garcia, Adrian|||0000-0002-3345-6093
  • Ruiz Ramírez, Marc|||0000-0001-6429-6347
  • Velasco Esteban, Luis Domingo|||0000-0002-7345-296X
Aiming at satisfying in a cost-effective manner the forecast traffic growth that future mobile networks will need to support, traditional distributed Radio Access Networks (RANs) are evolving towards centralized architectures. Specifically, the Cloud-RAN (C-RAN) architecture has shown that can alleviate to some extent the ever increasing Total Cost of Ownership in mobile networks. The current trend in C-RAN is to separate Remote Radio Heads (RRH) with radio frequency (RF) functions and Baseband Units (BBU) gathering baseband processing. This functional split allows keeping RF modules close to the antennas while placing BBUs at centralized locations so they can be shared among different sites and even be virtualized. However, some issues still need to be addressed in future mobile networks, especially due to the dynamicity of services and the strict constraints imposed by the interfaces needed. Since connectivity reconfiguration for backhaul interfaces (X2 and S1) needs to be provided as an all-or-nothing request to enable mobile resources reconfiguration in a geographical area, in this paper we propose dynamic Customer Virtual Network (CVN) reconfiguration to be supported in metro and core network segments. Additionally, such CVN requests must include Quality of Service constraints to ensure specific delay constraints, as well as bitrate guarantees to avoid service interruption. An efficient algorithm is presented for the CVN reconfiguration problem and exhaustive simulation results study its performance on realistic scenarios
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Fully disaggregated ROADM white box with NETCONF/YANG control, telemetry, and machine learning-based monitoring

  • Sgambelluri, Andrea
  • Izquierdo Zaragoza, José Luis|||0000-0003-4013-9712
  • Giorgetti, Alessio
  • Gifre Renom, Lluís
  • Velasco Esteban, Luis Domingo|||0000-0002-7345-296X
  • Paolucci, Francesco
  • Sambo, Nicola
  • Fresi, Francesco
  • Castoldi, Piero
  • Morro, Roberto
  • Riccardi, Emilio
  • D'Errico, Alessio
  • Cugini, Filippo
A first demonstration of ROADM White Box augmented with machine learning capabilities is demonstrated. The white box includes various level of disaggregation, NETCONF/YANG control, telemetry and spectrum-based advanced monitoring functionalities., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Feature-based optical spectrum monitoring for failure detection and identification

  • Shariati, Mohammad Behnam
  • Ruiz Ramírez, Marc|||0000-0001-6429-6347
  • Comellas Colomé, Jaume|||0000-0002-9129-0562
  • Velasco Esteban, Luis Domingo|||0000-0002-7345-296X
© 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works, In this paper, we explore the benefits of analysing the optical spectrum of lightpaths for soft-failure detection and identification in Spectrum Switched Optical Network. We present a framework exploiting machine learning (ML) based algorithms that uses descriptive models of the optical spectrum of a lightpath in different points along its route to detect whether the optical signal experiences anomalies reflecting a failure in the intermediate nodes. Our proposal targets the two most common filter-related soft-failures; filter shift (FS) and filter tightening (FT), which noticeably deform the expected shape of the optical spectrum. In this regard, filter cascading is a key challenge as it affects the shape of the optical spectrum similar to FT. Our proposals avoid the misclassification of properly operating signals when normal filter cascading effects is present. Extensive numerical results are presented to compare the performance of the proposed approaches in terms of accuracy and robustness., This work has received funding from the European Commission’s H2020 METRO-HAUL project (G.A. nº 761727), the AEI/FEDER TWINS project (TEC2017-90097-R), and ICREA Institution, Peer Reviewed
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Fully disaggregated ROADM white box with NETCONF/YANG control, telemetry, and machine learning-based monitoring

  • Sgambelluri, Andrea
  • Izquierdo Zaragoza, José Luis
  • Giorgetti, Alessio
  • Gifre Renom, Lluís
  • Velasco Esteban, Luis Domingo
  • Paolucci, Francesco
  • Sambo, Nicola
  • Fresi, Francesco
  • Castoldi, Piero
  • Morro, Roberto
  • Riccardi, Emilio
  • D'Errico, Alessio
  • Cugini, Filippo
A first demonstration of ROADM White Box augmented with machine learning capabilities is demonstrated. The white box includes various level of disaggregation, NETCONF/YANG control, telemetry and spectrum-based advanced monitoring functionalities., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Combining a machine learning and optimization for early pre-FEC BER degradation to meet committed QoS

  • Pérez Vela, Alba
  • Ruiz Ramírez, Marc
  • Cugini, Filippo
  • Velasco Esteban, Luis Domingo
Monitoring the physical layer is key to detect bit error rate (BER) degradation caused by failures and to identify the cause of the failure and localize the failed elements. Once the failure has been detected, actions can be taken to reduce as much as possible its impact on the network. Commercially available optical equipment are able to correct degraded optical signals by means of Forward Error Correction (FEC) algorithms. A value of pre-FEC BER over a pre-defined threshold would imply a non-error-free post-FEC transmission and, as a result, communication would be disrupted. Therefore, a prompt detection of lightpaths with excessive pre-FEC BER can help to greatly reduce such SLA violations, in particular when supporting vlinks. As a result of the above, it would be desirable to anticipate such degradations and apply re-optimization to re-route those affected demands according to their SLAs in order to reduce the affected traffic after the degradation is detected. Designing algorithms capable of promptly detect distinct BER anomaly patterns would be desirable. The objective would be to anticipate intolerable BER values as much as possible aiming at leaving enough time to plan a re-routing procedure during off-peak hours. In this paper, we propose an effective machine learning-based algorithm to localize and identify the most probable cause of failure impacting a given service, as well as a re-optimization algorithm to re-route affected demands, targeting at reducing SLA violation. Results show that the proposed detection and re-routing algorithms noticeably reduce bandwidth and number of demands affected., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Autonomous operations in optical networks

  • Pérez Vela, Alba
  • Ruiz Ramírez, Marc
  • Velasco Esteban, Luis Domingo
© 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., Machine Learning (ML) has already proven its benefits for network operation, being a sub-domain of artificial intelligence, it is highly suitable for complex system representation. In this paper, basic ML concepts are reviewed, as well as its integration into existing network control and management planes. Then, a use case focused on soft-failure detection is presented in detail covering optical spectrum analysis and ML algorithms; the technique relies on the widespread deployment of cost-effective optical spectrum analyzer (OSA). Finally, the retrieved optical parameters are analyzed using ML algorithms giving rise to illustrative results., The research leading to these results has received funding from the Spanish MINECO TWINS project (TEC2017-90097-R), from the EC through the METRO-HAUL project (G.A. no 761727), and from the Catalan Institution for Research and Advanced Studies (ICREA)., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Learning from the optical spectrum: soft-failure identification and localization

  • Velasco Esteban, Luis Domingo
  • Shariati, Mohammad Behnam
  • Vela, Alba P.
  • Comellas Colomé, Jaume
  • Ruiz Calvo, Manuel
The availability of coarse-resolution cost-effective Optical Spectrum Analyzers (OSA) allows its widespread deployment in operators' networks. In this paper, several machine learning approaches for failure identification and localization that take advantage of OSAs are presented, Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Adaptation of the residual signal for filter failure detection in scenarios with multiple filter types

  • Barzegar, Sima
  • Ruiz Ramírez, Marc
  • Velasco Esteban, Luis Domingo
© 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., To monitor the optical spectra of outgoing links, Optical Spectrum Analyzers (OSA) can deployed in the optical nodes acquiring the optical spectra of outgoing links and analyzing optical signals to detect those soft-failures that affect the shape of the spectrum, like Filter Shift and Filter Tightening. In this paper, we rely on the residual-based approach spectra analysis that uses a set of residual signals calculated by subtracting the measured signal by the OSA from an expected version of the signal. In this paper, we assume a network scenario, where nodes with optical filters of different characteristics (e.g., 2nd, 3rd, and 4th order Gaussian) are installed. In this scenario, the possibility of generating complete data set for machine learning training purposes that include every combination of filter and failure is difficult to achieve. Instead, we propose an adaptation mechanism that transform the expected effects of every filter type thus, emulating single filter-type networks. Exhaustive experiments are carried out to show the performance of the proposed adaptation under filter failures., This work was partially supported by the EC through the METRO-HAUL project (G.A. nº 761727), from the AEI/FEDER TWINS project (TEC2017-90097-R), and from the Catalan ICREA Institution., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Monitoring and data analytics-triggered reconfiguration in partially disaggregated optical networks

  • Gifré Renom, Lluís
  • Boitier, Fabien
  • Delezoide, Camille
  • Ruiz Ramírez, Marc
  • Buffa, Marta
  • Morea, Annalisa
  • Casellas Regi, Ramon
  • Velasco Esteban, Luis Domingo
  • Layec, Patricia
©2020 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., We present ADONIS, which stands for Aggregator/Disaggregator for Optical Network equIpmentS, a novel open device agent able to construct logical network devices from (dis)aggregation of physical components in order to expose meaningful network devices to the SDN controller. We experimentally assess it by means of a control closed-loop involving ADONIS, a Software Defined Network controller, a Monitoring and Data Analytics system, and a novel reconfiguration tool, SMART-A., The research leading to these results has received funding from the EC through the METRO-HAUL (G.A. nº 761727)., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

On the benefits of elastic spectrum management in multi-hour filterless metro networks

  • Pedreno Manresa, J.J.
  • Izquierdo Zaragoza, José Luis
  • Cugini, Filippo
  • Pavón Mariño, Pablo
The dawn of 5G is pushing operators to deploy high-capacity, agile networks capable of adapting to time-varying traffic patterns, especially into metro sections. ROADMs are key enablers for agility in the optical layer, however the benefits of this agility do not always compensate for increased costs. As such, filterless optical networks are emerging as a cost-effective and reliable solution compared to active photonics, thanks to a winning combination of coherent transponders and passive splitters/couplers. However, spectrum allocation management policies are of paramount importance to maximize the overall network throughput. In this paper, we focus on a filterless metro network where the hourly variation of the demands traffic is known, coming from historic data estimations. Then, we observe how the knowledge of the traffic profiles can be exploited. To assess this, we evaluate the performance, in terms of throughput, of three different spectrum management approaches: (i) fixed, where lightpaths remain static along time once allocated; (ii) semi-elastic, where lightpath-bandwidth vary according to current traffic requirements, but central frequency remains fixed; and (iii) hitless full-elastic, where any lightpath parameter may be reconfigured without disrupting the traffic. Besides, we consider two transponder types equipped with (i) shared or (ii) independent tunable lasers for transmission and reception, which affects to spectrum allocation of bidirectional connections. According to our results, the semi-elastic approach clearly outperforms the fixed approach (23-33% more throughput) with a reduced gap to the hitless full-elastic case (10-24% less throughput), especially considering that the latter is not commercially available yet. Interestingly, using dual-laser transponders only yields a 10% gain with respect to single-laser transponders for the semi-elastic scenario, and thus may not justify the extra hardware., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Flexible fog computing and telecom architecture for 5G networks

  • Velasco Esteban, Luis Domingo
  • Ruiz Ramírez, Marc
We review a novel, secure, highly distributed and ultra-dense fog computing infrastructure, which can be allocated at the extreme edge of a wired/wireless network for a Telecom Operator to provide multiple unified, cost-effective and new 5G services, such as Network Function Virtualization (NFV), Mobile Edge Computing (MEC), and services for third parties (e.g., smart cities, vertical industries or Internet of Things (IoT)). The distributed and programmable fog technologies are expected to strengthen the position of the Mobile Network and cloud markets; key benefits are the dynamic deployment of new distributed low-latency services. The architecture consists of three main building blocks: a) a scalable node, that is seamlessly integrated in the Telecom infrastructure; b) a controller, focused on service assurance, that is integrated in the management and orchestration architecture of the Telecom operator; and c) services running on top of the Telecom infrastructure., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Autonomic NFV network services on top of disaggregated optical metro networks

  • Esmenats, P. R.
  • Casellas Regi, Ramon
  • Gifre Renom, Lluís
  • Pérez Vela, Alba
  • Ruiz Ramírez, Marc
  • Martínez, Ricardo
  • Velasco Esteban, Luis Domingo
Control and orchestration of metro network to support dynamic provisioning and reconfiguration of services based on Virtual Network Functions are demonstrated through a CDN example. Interaction among CASTOR MDA, OSM and ONOS is exhibited., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Experimental assessment of SDN-enabled reconfigurable OPSquare data center networks with QoS guarantees

  • Xue, Xuwei
  • Wang, Fulong
  • Agraz Bujan, Fernando
  • Pagès Cruz, Albert
  • Pan, B
  • Yan, F
  • Spadaro, Salvatore
  • Calabretta, Nicola
An SDN-controlled DCN enabling QoS-driven network-slice reconfiguration and packet priority updating is experimentally assessed. Network-slice can be reconfigured within 150 ms to decrease packet loss. The measured ToR-to-ToR latency of high-priority packets is 260.55 ns., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Soft-failure localization and device working parameters estimation in disaggregated scenarios

  • Barzegar, Sima
  • Virgillito, E.
  • Ruiz Ramírez, Marc
  • Ferrari, A.
  • Napoli, Antonio
  • Curri, Vittorio
  • Velasco Esteban, Luis Domingo
A soft-failure localization and key working parameters estimation system is proposed for network diagnosis and maintenance. We show that a double analysis of monitoring data and estimated working parameters greatly anticipates degradations, The research leading to these results has received funding from the EC through the METRO-HAUL project (G.A. nº 761727), from the Spanish MINECO TWINS project (TEC2017-90097-R), and from the Catalan Institution for Research and Advanced Studies (ICREA), Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Collective self-learning by exchanging ML models

  • Ruiz Ramírez, Marc
  • Boitier, Fabien
  • Layec, Patricia
  • Velasco Esteban, Luis Domingo
Collective self-learning based on Machine Learning (ML) model sharing and combination is proposed to accelerate ML-based algorithm deployment. The considered architecture is presented, together with different alternatives for combining ML models. Performance analysis is carried out on an illustrative use case for autonomic optical transmission., The research leading to these results has received funding from the AEI/FEDER TWINS project (TEC2017-90097-R), from the EC METRO-HAUL project (G.A. nº 761727), and from the Catalan ICREA Institution., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Allocating multi-rate requests to single-rate lightpaths in Flexgrid optical networks

  • Comellas Colomé, Jaume
  • Junyent Giralt, Gabriel
Some ideas about the effect of appropriately selecting the size of the connections conveyed by multi-rate flexgrid optical networks are discussed. The main focus is on illustrating the relation between spectrum consumption and effective carried traffic. It is demonstrated that by introducing some order in the spectrum use, so limiting the network flexibility, its global performance is notably improved. The main contribution consists in demonstrating that using single sized lightpaths, the network throughput at any given blocking probability value increases compared to the multi-rate case. This conclusion is supported by extensive simulations under dynamic traffic conditions, considering different network topologies and traffic profiles., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Experimental assessment of big data-backed video distribution in the telecom cloud

  • Gifre Renom, Lluís
  • Ruiz Ramírez, Marc
  • Velasco Esteban, Luis Domingo
The telecom infrastructure is undergoing a huge transformation since operators are deploying their own cloud infrastructure to provide cloud services and enabling Network Functions Virtualization (NFV). The resulting infrastructure is referred to as the telecom cloud. NFV decouples network functions from proprietary hardware appliances, so they can be implemented in software and deployed on virtual machines (VM) running on commercial off-the-shelf computing hardware. One of the network functions currently being considered by many operators is Content Delivery Network (CDN) for live-TV and Video on Demand (VoD) distribution. In fact, a significant advance in video delivery is the standardized MPEG Dynamic Adaptive Streaming over HTTP (MPEG-DASH) technique that enables media content delivering over the Internet using standard HTTP web server infrastructure. Besides, the CDN reconfiguration, motivated by an increase in the number of users demanding content, entails a Virtual Network Topology (VNT) reconfiguration to cope with the demanded connection capacities. A Big Data-based CDN manager, fulfilling the ETSI NFV guidelines, is proposed to adapt the virtualized CDN function to current and future demand. Its feasibility to control virtualized components while collecting monitoring data is experimentally assessed in a real environment., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Enabling data analytics and machine learning for 5G services within disaggregated multi-layer transport networks

  • Casellas Regi, Ramón
  • Martínez Rivera, Ricardo Victor
  • Velasco Esteban, Luis Domingo
  • Vilalta, Ricard
  • Pavón Mariño, Pablo
  • King, Daniel
  • Muñoz González, Raül
Recent advances, related to the concepts of Artificial Intelligence (AI) and Machine Learning (ML) and with applications across multiple technology domains, have gathered significant attention due, in particular, to the overall performance improvement of such automated systems when compared to methods relying on human operation. Consequently, using AI/ML for managing, operating and optimizing transport networks is increasingly seen as a potential opportunity targeting, notably, large and complex environments.Such AI-assisted automated network operation is expected to facilitate innovation in multiple aspects related to the control and management of future optical networks and is a promising milestone in the evolution towards autonomous networks, where networks self-adjust parameters such as transceiver configuration.To accomplish this goal, current network control, management and orchestration systems need to enable the application of AI/ML techniques. It is arguable that Software-Defined Networking (SDN) principles, favouring centralized control deployments, featured application programming interfaces and the development of a related application ecosystem are well positioned to facilitate the progressive introduction of such techniques, starting, notably, in allowing efficient and massive monitoring and data collection.In this paper, we present the control, orchestration and management architecture designed to allow the automatic deployment of 5G services (such as ETSI NFV network services) across metropolitan networks, conceived to interface 5G access networks with elastic core optical networks at multi Tb/s. This network segment, referred to as Metro-haul, is composed of infrastructure nodes that encompass networking, storage and processing resources, which are in turn interconnected by open and disaggregated optical networks. In particular, we detail subsystems like the Monitoring and Data Analytics or the in-operation planning backend that extend current SDN based network control to account for new use cases., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Data analytics based origin-destination core traffic modelling

  • Morales Alcaide, Fernando
  • Ruiz Ramírez, Marc
  • Velasco Esteban, Luis Domingo
Traffic monitoring is an essential task for network operators since it allows evaluating network performance. Monitoring data from origin-destination (OD) traffic in core virtual network topologies can be collected from packet nodes and stored in a repository for further analysis, e.g., to detect anomalies or to create predicted traffic matrices for the near future. In this paper we propose a set of modules to support data analytics-based algorithms along with a machine learning procedure based on artificial neural networks (ANN) that provides robust and adaptive traffic models., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Quantifying dynamic traffic in elastic optical networks

  • Comellas Colomé, Jaume
  • Vicario, L.
  • Junyent Giralt, Gabriel
This work is focused on the characterization of the traffic allocated in Elastic Optical backbone networks when different dynamic traffic profiles are generated. Extensive simulations are done to statistically model the real-time traffic carried by this kind of networks. Well known topologies such as NSFNet and US_Backbone are considered as well as different parameters applied to the traffic generation. Some interesting conclusions are extracted about the properties of the traffic supported by Elastic Optical Networks under dynamic conditions., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Monitoring and data analytics for optical networking

  • Velasco Esteban, Luis Domingo
  • Ruiz Ramírez, Marc
© 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., Increased levels of automation will be necessary in view of more stringent performance requirements that next generation optical transport networks need to support in a near term, not only for high capacity, but, even more importantly, for dynamicity, latency, and availability. All these aspects will become more relevant with the growing complexity of modern networks. Network automation targets resource re-optimization to rapidly adapt the network to the expected conditions, quick degradation detection to improve the quality of the connections, as well as failure detection and identification to facilitate maintenance. Network automation requires and implies the collection of data for performance monitoring, being then elaborated by data analytics algorithms to produce meaningful inputs for the network controller, which will finally program the underlying devices. In this paper, we analyze alternative architectures for monitoring and data analytics (MDA) and illustrative control loops are presented aiming at validating the usefulness of MDA to automate optical networks operation., The research leading to these results has received funding from the EC through the METRO-HAUL project (G.A. nº 761727), from the Spanish MINECO TWINS project (TEC2017-90097-R), and from the Catalan Institution for Research and Advanced Studies (ICREA)., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Impact of traffic delay tolerance on elastic optical networks performance

  • Comellas Colomé, Jaume
  • Nourmohammadi, Farzaneh
  • Junyent Giralt, Gabriel
© 2020 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., Elastic Optical Networks (EON) are considered as a valuable solution to enhance optical networks efficiency due to their superior network spectrum use. By breaking the traditional fixed grid of WDM networks, the spectrum assigned to connection demands is adapted to their requirements. On the other hand, connection parameters such as baud-rate or modulation format are dynamically adjusted taking into account the current network conditions. Therefore, EONs fit perfectly with the heterogeneous characteristics of current and future network traffic loads. Nevertheless, when EONs are in operation, they present some weaknesses mainly derived from the spectrum fragmentation generated by the dynamic establishment and torn-down of these diverse connections. The randomness of connections generation results in the blocking of some connections even when the total load supported by the network is not so high. In this work, we analyse how traffic tolerance to delay impacts on the EON performance, and present simulation obtained results when connection demands have no strict delay constraints compared with demands which need to be real-time served., This work was partially supported by the EC through the METRO-HAUL project (G.A. 761727), and from the Spanish MINECO TWINS project (TEC2017-90097-R)., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Applying data visualization for failure localization

  • Vela, Alba P.
  • Ruiz Ramírez, Marc
  • Velasco Esteban, Luis Domingo
Data visualization is applied to BER measures in a bigdata repository. Bubble charts are produced to identify lightpaths with increasing BER and spectrum color maps are then used to identify the most likely degraded link, Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Bringing data analytics to the network nodes for efficient traffic anomalies detection

  • Pérez Vela, Alba
  • Ruiz Ramírez, Marc
  • Velasco Esteban, Luis Domingo
Traffic anomalies can create network congestion, so its prompt and accurate detection would allow network operators to make decisions to guarantee the network performance avoiding services to experience any perturbation. In this paper, we focus on origin-destination (OD) traffic anomalies; to efficiently detect those, we study two different anomaly detection methods based on data analytics and combine them with three monitoring strategies. In view of the short monitoring period needed to reduce anomaly detection, which entails large amount of monitoring data to be collected and analyzed in a centralized repository, we propose bringing data analytics to the network nodes to efficiently detect traffic anomalies, while keeping traffic estimation centralized. Exhaustive simulation results on a realistic network scenario show that the monitoring period should be as low as possible (e.g., 1 min) to keep anomaly detection times low, which clearly motivates to place traffic anomaly detection function in the network nodes., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Adapting the virtual network topology to near future traffic

  • Morales Alcaide, Fernando
  • Festa, Paola
  • Ruiz Ramírez, Marc
  • Velasco Esteban, Luis Domingo
The introduction of new services requiring large and dynamic bitrate connectivity can cause changes in the direction of the traffic in metro and even core network segments along the day. This leads to large overprovisioning in statically managed virtual network topologies (VNT), designed to cope with the traffic forecast. To reduce expenses while ensuring the required grade of service, in this paper we propose the VNT reconfiguration approach based on current and near-future traffic matrices (VENTURE) to regularly adapt the topology to both, the current and future traffic volume and direction. The problem is formally stated and a heuristic algorithm is proposed to solve it., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Distributed and centralized options for self-learning

  • Tabatabaeimehr, Fatemehsadat
  • Ruiz Ramírez, Marc
  • Velasco Esteban, Luis Domingo
© 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., In general, the availability of enough real data from real fog computing scenarios to produce accurate Machine Learning (ML) models is rarely ensured since new equipment, techniques, etc., are continuously being deployed in the field. Although an option is to generate data from simulation and lab experiments, such data could not cover the whole features space, which would translate into ML models inaccuracies. In this paper, we propose a self-learning approach to facilitate ML deployment in real scenarios. A dataset for ML training can be initially populated based on the results from simulation and lab experiments and once ML models are generated, ML re-training can be performed after inaccuracies are detected to improve their precision. Illustrative numerical results show the benefits from the proposed self-learning approach for two general use cases of regression and classification., This work was partially supported by the EC through the METRO-HAUL project (G.A. nº 761727), from the AEI/FEDER TWINS project (TEC2017-90097-R), and from the Catalan ICREA Institution., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Optical signal tracking for robust PAM4 deployment in filterless metro network scenarios

  • Shariati, Mohammad Behnam
  • Fresi, Francesco
  • Ruiz Ramírez, Marc
  • Cugini, Filippo
  • Velasco Esteban, Luis Domingo
© 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., Highly accurate and reliable optical signal tracking is proposed that estimates sub-GHz laser drift failures by analyzing spectra acquired by cost-effective coarse-granular OSAs. Its application on PAM4 systems in filterless metro networks brings added robustness., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Soft failure localization in elastic optical networks

  • Vela, A.P.
  • Shariati, Mohammad Behnam
  • Ruiz Ramírez, Marc
  • Comellas Colomé, Jaume
  • Velasco Esteban, Luis Domingo
Soft failure localization to early detect service level agreement violations is of paramount importance in elastic optical networks (EONs), while it allows anticipating possible hard failure events. Nowadays, effective and automated solutions for soft failure localization during lightpaths' commissioning testing and operation are still missing. In this paper, we focus on presenting soft failure localization algorithms based on two different active monitoring techniques. First, the Testing optIcal Switching at connection SetUp timE (TISSUE) algorithm is proposed to localize soft failures during commissioning testing phase by elaborating the estimated bit-error rate (BER) values provided by low-cost optical testing channel (OTC) modules. Second, the FailurE causE Localization for optIcal NetworkinG (FEELING) algorithm is proposed to localize failures during lightpath operation using cost-effective optical spectrum analyzers (OSAs) widely deployed in network nodes. Results are presented to validate both algorithms in the event of several soft failures affecting lasers and filters. © 2018 IEEE., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Periodic defragmentation in elastic optical networks

  • Comellas Colomé, Jaume
  • Vicario, L.
  • Junyent Giralt, Gabriel
The main weakness of Elastic Optical Networks (EON) stems from spectrum fragmentation. An analysis of periodic defragmentation in EONs under dynamic traffic conditions is carried out. The effects of different defragmentation parameters on the EON performance are evaluated, and appropriate values, guaranteeing suitable network performance while keeping the network control complexity at a reasonable point, are obtained through simulations. Different network topologies as well as traffic conditions are simulated to assess the validity of the obtained results., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Interfaces for monitoring and data analytics systems

  • Gifré Renom, Lluís
  • Ruiz Ramírez, Marc
  • Velasco Esteban, Luis Domingo
© 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., Software defined networks (SDN) represents one of the most relevant innovations in recent years for the telecom industry. Major operators are planning to progressively migrate their transport networks, including optical, to such paradigm to take advantage of the programmability of connectivity services. In SDN, resources are abstracted at the networking level and exposed via a north-bound interface to management systems. On the other hand, Monitoring and Data Analytics (MDA) systems can collect monitoring data and make them accessible to external systems, such as billing platforms, to accurately charge users for the services they consume. In addition, in dynamic scenarios, where resources could be created, modified and removed on demand, MDA systems can also facilitate autonomic networking, and thus coordination between SDN and MDA is strictly required. In this paper, we present two monitoring interfaces for MDA systems: the Monitoring (M)-Control, Orchestration, and Management (COM) interface, an east-west interface that enables MDA systems to gather and synchronize operational and telemetry information from multiple COM modules, e.g., an SDN controller or a cloud resource manager, and issue recommendations to the COM modules as a result of data analysis; and the IO4 interface, a MDA north-bound interface that facilitates exporting monitored data to external systems. Functionality of each interface is described from the conceptual viewpoint and relevant operational workflows are proposed., This work was partially supported by the EC through the METRO-HAUL project (G.A. nº 761727), from the AEI/FEDER TWINS project (TEC2017-90097-R), and from the Catalan ICREA Institution., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Visualization tools for enhancing failure localization [Invited]

  • Pérez Vela, Alba
  • Ruiz Ramírez, Marc
  • Velasco Esteban, Luis Domingo
© 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., The use of data visualization techniques is explored to guide human operators in failure localization tasks. In this paper, we consider the case where a subset of lightpaths’ experience gradual BER degradation. We aim at using visualization techniques to find out the most likely resource that is responsible for such degradation, as well as the start of the degradation. Note that the challenge is to analyze an amount of data that might be huge (e.g., 1 sample every 15 min for the last, let us say, 4 weeks, for every lightpath in the network) and display charts in seconds, while revealing meaningful information for human operators. The proposed task-oriented visualization tackles such challenge by using charts that first help finding the subset of affected lightpaths thus, reducing the number of lightpaths to be analyzed. In a second step, different charts are used to reduce the element in failure as much as possible. Finally, the few suspicious elements can be analyzed in detail using more traditional timeline graphs., The research leading to these results has received funding from the Spanish MINECO TWINS project (TEC2017-90097-R), from the EC through the METRO-HAUL project (G.A. no 761727), and from the Catalan Institution for Research and Advanced Studies (ICREA)., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

CASTOR: An architecture to bring cognition to transport networks

  • Velasco Esteban, Luis Domingo
  • Gifre Renom, Lluís
  • Izquierdo Zaragoza, José Luis
  • Julián Moreno, Guillermo
  • López de Vergara Méndez, Jorge E.
CASTOR architecture to enable cognitive networking is demonstrated. Extended nodes make local decisions, whilst a centralized system beside the network controller makes network-wide decisions. Interaction with ONOS, Net2Plan, and passive monitoring devices is exhibited., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Toward multilayer disaggregated node telemetry and local decision making

  • Velasco Esteban, Luis Domingo
  • Gifre Renom, Lluís
  • Izquierdo Zaragoza, José Luis
A generic node agent supporting disaggregated node telemetry is presented. Data collection close to devices enable making local decisions, leveraging SDN controllers for network-wide operations. The agent is demonstrated in a BER-triggered transponder reconfiguration scenario., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Monitoring and data analytics: Analyzing the optical spectrum for soft-failure detection and identification

  • Shariati, Mohammad Behnam
  • Vela, Alba P.
  • Ruiz Ramírez, Marc
  • Velasco Esteban, Luis Domingo
Failure detection is essential in optical networks as a result of the huge amount of traffic that optical connections support. Additionally, the cause of failure needs to be identified so failed resources can be excluded from the computation of restoration paths. In the case of soft-failures, their prompt detection, identification, and localization make that recovery can be triggered before excessive errors in optical connections translate into errors on the supported services or even become disrupted. Therefore, Monitoring and Data Analytics (MDA) become of paramount importance in the case of soft-failures. In this paper, we review a MDA architecture that reduces remarkably detection and identification times, while facilitating failure localization. In addition, we rely on Optical Spectrum Analyzers (OSA) deployed in the optical nodes as monitoring devices acquiring the optical spectrum of outgoing links. Analyzing the optical spectrum of optical connections, specific soft-failures that affect the shape of the spectrum can be detected. A workflow consisting of machine learning algorithms, designed to be integrated in the aforementioned MDA architecture, will be studied to analyze the optical spectrum of a given optical connection acquired in a node and to determine whether a filter failure is affecting it, and in such case, what is the type of filter failure and its magnitude. Exhaustive results are presented allowing to evaluate the proposed method. © 2018 IFIP., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Using dual-path allocation for partial traffic protection in elastic optical networks

  • Junyent Giralt, Gabriel
  • Comellas Colomé, Jaume
Multi-path allocation has been proposed as a simple way to achieve traffic protection in different networking domains. In this work dual-path allocation is proposed for Elastic Optical Networks (EON) assuming dynamic offered traffic, and its performance is quantified by simulation. Indeed, dual-path allocation is compared with partial dedicated path protection assuming different scenarios. In addition to a qualitative comparison where advantages and drawbacks of both schemes are analysed, quantitative results for different backbone network scenarios are presented., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Smart filterless optical networks based on optical spectrum analysis

  • Ruiz Ramírez, Marc
  • Sgambelluri, Andrea
  • Cugini, Filippo
  • Velasco Esteban, Luis Domingo
© 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., Dynamic network operations can produce power fluctuations of the established connections in filterless optical networks. In addition, the gridless nature of filterless networks make that some (un)intentional effects such as transponders laser drift might disrupt the proper operation of lightpaths. To overcome these issues, we present a monitoring system exploiting data analytics and cost-effective optical spectrum analyzers to achieve smart filterless network operation. Experimental measurements are used to validate the proposed data analytics-based approaches, as well as to find the optimal resolution to achieve maximum performance with minimum cost., This work was partially supported by the EC through the METRO-HAUL project (G.A. nº 761727), from the AEI/FEDER TWINS project (TEC2017-90097-R), and from the Catalan ICREA Institution., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Real-time spectrum surveillance in filterless optical networks

  • Shariati, Mohammad Behnam
  • Ruiz Ramírez, Marc
  • Sgambelluri, Andrea
  • Cugini, Filippo
  • Velasco Esteban, Luis Domingo
A monitoring system exploiting data analytics and cost-effective optical spectrum analyzers with under 1.2GHz resolution is proposed. Its performance is demonstrated in a filterless network experimental test-bed. The system enables prompt action before lightpath disruption. © 2018 OSA., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Performance evaluation of the VNT reconfiguration algorithm based on traffic prediction

  • Morales Alcaide, Fernando
  • Festa, Paola
  • Ruiz Ramírez, Marc
  • Velasco Esteban, Luis Domingo
In a previous work, the Virtual Network Topology Reconfiguration problem based on Traffic Prediction (VENTURE) was proposed as a means of efficiently adapting core virtual network topology (VNT) to the near-future traffic. Although the benefits obtained by the VENTURE algorithm compared to using a purely reactive VNT reconfiguration approach seems to be clear, margin for improvement still remain and alternative solving methods for the VENTURE problem need to be considered. In this paper, the original VENTURE algorithm is compared against two state-of-the-art metaheuristics for combinatorial network optimization. The two metaheuristics are first presented and then adapted for the VENTURE problem. Finally, the performance of the VENTURE algorithm and the two proposed metaheuristics is numerically evaluated using an exact solving method as reference. © 2018 IEEE., Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Dynamic virtual network connectivity for C-RAN backhauling

  • Asensio Garcia, Adrian
  • Ruiz Ramírez, Marc
  • Velasco Esteban, Luis Domingo
Aiming at satisfying in a cost-effective manner the forecast traffic growth that future mobile networks will need to support, traditional distributed Radio Access Networks (RANs) are evolving towards centralized architectures. Specifically, the Cloud-RAN (C-RAN) architecture has shown that can alleviate to some extent the ever increasing Total Cost of Ownership in mobile networks. The current trend in C-RAN is to separate Remote Radio Heads (RRH) with radio frequency (RF) functions and Baseband Units (BBU) gathering baseband processing. This functional split allows keeping RF modules close to the antennas while placing BBUs at centralized locations so they can be shared among different sites and even be virtualized. However, some issues still need to be addressed in future mobile networks, especially due to the dynamicity of services and the strict constraints imposed by the interfaces needed. Since connectivity reconfiguration for backhaul interfaces (X2 and S1) needs to be provided as an all-or-nothing request to enable mobile resources reconfiguration in a geographical area, in this paper we propose dynamic Customer Virtual Network (CVN) reconfiguration to be supported in metro and core network segments. Additionally, such CVN requests must include Quality of Service constraints to ensure specific delay constraints, as well as bitrate guarantees to avoid service interruption. An efficient algorithm is presented for the CVN reconfiguration problem and exhaustive simulation results study its performance on realistic scenarios, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

CASTOR: a monitoring and data analytics architecture to support autonomic domain and slice networking

  • Vela, A.P.
  • Gifre Renom, Lluís
  • González de Dios, Oscar
  • Ruiz Ramírez, Marc
  • Velasco Esteban, Luis Domingo
Network slices combine resource virtualization with the isolation level required by future 5G applications. In addition, the use of Monitoring and Data Analytics (MDA) helps to maintain the required network performance, while reducing total cost of ownership. In this paper, we present CASTOR, an architecture to enable autonomic domain and slice networking. MDA agents use data analytics to make local decisions close to network devices, whereas MDA controllers collate and export metered data transparently to customer controllers, all of them leveraging customizable and isolated data analytics processes. Discovered knowledge can be applied for both proactive and reactive network slice reconfiguration, triggered either by service providers or customers, thanks to the interaction with state-of-the-art software-defined networking controllers and planning tools. © 2018 IEEE., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Spectrally and spatially flexible optical networks: recent developments and findings

  • Shariati, Mohammad Behnam
  • Klonidis, Dimitrios
  • Marom, Dan M.
  • Comellas Colomé, Jaume
  • Velasco Esteban, Luis Domingo
  • Tomkos, Ioannis
Space division multiplexing (SDM) has been proposed as the ultimate solution to address the capacity crunch of optical transport networks. The efficient utilization of SDM requires some forms of spatial integration which is expected to bring huge cost savings for the deployment of SDM-based optical networks. Spatially integrated components with different characteristics have been largely explored and demonstrated for the realization of SDM networks, including; transmission media, transceivers with sophisticated digital signal processing (DSP) units, amplifiers, and ultimately optical switching nodes which are the key elements for the realization of transparent optical networks exploiting SDM technologies. As a consequence, in contrast to the currently deployed optical networks based on standard single mode fibers (SSMF), the next generation of optical networks exploiting SDM technologies can be realized utilizing various kinds of transmission media and the other corresponding elements. However, due in part to the physical properties of different components, their complexity, and the technology limitations not all of them meet the economic feasibility and a desired level of network-wide performance. Therefore, careful analysis should be done revealing the most appropriate and cost-effective solutions. In this work, we review the recent developments and findings which pave the way for a simplified and efficient implementation of the spectrally and spatially flexible optical networks., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Out-of-field generic ML training with in-field specific adaptation to facilitate ML deployments

  • Shariati, Mohammad Behnam
  • Ruiz Ramírez, Marc
  • Velasco Esteban, Luis Domingo
© 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., A two-phase strategy to facilitate ML algorithm deployment in real networks is demonstrated: out-of-field training uses data from simulation and testbed experiments with generic equipment whereas in-field adaptation is applied to support heterogeneous equipment., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Autonomic content delivery network service

  • Velasco Esteban, Luis Domingo
  • Gifré Renom, Lluís
  • Ruiz Ramírez, Marc
© 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., We focus on a use case where a virtualized Content Delivery Network (CDN) service autonomously adapts to the load by requesting the instantiation of new VMs in selected leaf cache nodes, as well as by incrementing the capacity of the network connecting users with the caches. Autonomous decisions are supported by the use of data analytics techniques on monitoring data. Experimental validation supports the feasibility of the proposed autonomic CDN service., This work was partially supported by the EC through the METRO-HAUL project (G.A. nº 761727), from the AEI/FEDER TWINS project (TEC2017-90097-R), and from the Catalan Institution for Research and Advanced Studies (ICREA)., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Demonstration of monitoring and data analytics-triggered reconfiguration in partially disaggregated optical networks

  • Gifré Renom, Lluís
  • Boitier, Fabien
  • Delezoide, Camille
  • Ruiz Ramírez, Marc
  • Buffa, Marta
  • Morea, Annalisa
  • Casellas Regi, Ramon
  • Velasco Esteban, Luis Domingo
  • Layec, Patricia
©2020 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., We demonstrate a novel agent for optical disaggregated optical networks. When the Monitoring and Data Analytics detects a degradation, it recommends the SDN controller to trigger a network reconfiguration computed by a novel planning tool., The research leading to these results has received funding from the EC through the METRO-HAUL (G.A. nº 761727)., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

A latency-aware real-time video surveillance demo: network slicing for improving public safety

  • Shariati, Mohammad Behnam
  • Pedreno Manresa, José Juan
  • Dochhan, Annika
  • Siddique Muqaddas, Abubakar
  • Casellas Regi, Ramon
  • González de Dios, Oscar
  • Luque Canto, L.
  • Lent, Bodo
  • López de Vergara Méndez, Jorge Enrique
  • Velasco Esteban, Luis Domingo
We report the automated deployment of 5G services across a latency-aware, semi- disaggregated, and virtualized metro network. We summarize the key findings in a detailed analysis of end-to-end latency, service setup time, and soft-failure detection time., The research leading to these results has received funding from the EC and BMBF through the METRO-HAUL project (G.A. No. 761727) and OTB-5G+ project (reference No. 16KIS0979K)., Peer Reviewed, Article signat per 24 autors/es: B. Shariati, Fraunhofer HHI, Germany / J. J. Pedreno-Manresa, ADVA, Germany / A. Dochhan, ADVA, Germany / A. S. Muqaddas, UOB, UK / R. Casellas, CTTC/CERCA, Spain / O. González de Dios, Telefónica, Spain / L. L. Canto, Telefónica, Spain / B. Lent, Qognify GmbH, Germany / J. E. López de Vergara, Naudit HPCN, Spain / S. López-Buedo, Naudit HPCN, Spain / F. J. Moreno, UPCT, Spain / P. Pavón, UPCT, Spain / L. Velasco, UPC, Spain / S. Patri, ADVA, Germany / A. Giorgetti, CNIT, Italy / IEIIT-CNR, F. Cugini CNIT, Italy / A. Sgambelluri, CNIT, Italy / R. Nejabati, UOB, UK / D. Simeonidou, UOB, UK / R.-P. Braun, Deutsche Telekom, Germany / A. Autenrieth, ADVA, Germany / J.-P. Elbers, ADVA, Germany / J. K. Fischer, Fraunhofer HHI, Germany / R. Freund, Fraunhofer HHI, Germany, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Automated end to end Carrier Ethernet provisioning over a disaggregated WDM metro network with a hierarchical SDN control and monitoring platform

  • Morro, Roberto
  • Lucrezia, Francesco
  • Gomes, P.
  • Casellas Regi, Ramón
  • Giorgetti, Alessio
  • Velasco Esteban, Luis Domingo
  • Riccardi, Emilio
  • Chiadò Piat, Anna
  • Percelsi, Alessandro
  • Pedro, João
  • Gifre Renom, Lluís
  • Sgambelluri, Andrea
  • Risso, Fulvio
  • Marchetto, Guido
This demo shows how a hierarchical control plane of ONOS SDN controllers orchestrates the dynamic provisioning of end-to-end Carrier Ethernet circuits on a composite network, programming the whole data path from the CPE to the core optical equipment., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Generation and analysis of service-based traffic flows

  • Bernal Escribano, Álvaro
  • Ruiz Ramírez, Marc
  • Velasco Esteban, Luis Domingo
© 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., The rapid availability of new services makes that network operators cannot exhaustively test their impact on the network or anticipate any capacity exhaustion. This situation will be worse with the imminent introduction of the 5G technology and the kind of totally new services that it will support. In this paper, we present CURSA-SQ, a methodology to analyze the network behavior when the specific traffic that would be generated by groups of service consumers is injected. CURSA-SQ includes input traffic flow modelling with second and sub-second granularity based on specific service and consumer behaviors. The methodology allows to accurately study traffic flows at the input and outputs of complex scenarios with multiples queues systems, as well as other metrics such as delays., This work was partially supported by the EC through the METRO-HAUL project (G.A. nº 761727), from the AEI/FEDER TWINS project (TEC2017-90097-R), and from the Catalan ICREA Institution., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Fault management based on machine learning

  • Velasco Esteban, Luis Domingo
  • Rafique, Danish
© 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., Machine Learning (ML) brings many benefits for network operation. In this paper, basic ML concepts and its integration into existing network control and management planes are reviewed. Case studies covering fault management are illustrated., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Feature-based optical spectrum monitoring for failure detection and identification

  • Shariati, Mohammad Behnam
  • Ruiz Ramírez, Marc
  • Comellas Colomé, Jaume
  • Velasco Esteban, Luis Domingo
© 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works, In this paper, we explore the benefits of analysing the optical spectrum of lightpaths for soft-failure detection and identification in Spectrum Switched Optical Network. We present a framework exploiting machine learning (ML) based algorithms that uses descriptive models of the optical spectrum of a lightpath in different points along its route to detect whether the optical signal experiences anomalies reflecting a failure in the intermediate nodes. Our proposal targets the two most common filter-related soft-failures; filter shift (FS) and filter tightening (FT), which noticeably deform the expected shape of the optical spectrum. In this regard, filter cascading is a key challenge as it affects the shape of the optical spectrum similar to FT. Our proposals avoid the misclassification of properly operating signals when normal filter cascading effects is present. Extensive numerical results are presented to compare the performance of the proposed approaches in terms of accuracy and robustness., This work has received funding from the European Commission’s H2020 METRO-HAUL project (G.A. nº 761727), the AEI/FEDER TWINS project (TEC2017-90097-R), and ICREA Institution, Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Self-learning approaches for real optical networks

  • Ruiz Ramírez, Marc
  • Boitier, Fabien
  • Layec, Patricia
  • Velasco Esteban, Luis Domingo
© 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., Self-learning approaches to facilitate the deployment of ML algorithms in real networks are analyzed and their performance evaluated through an illustrative use case. Results show large benefits of collective self-learning with centralized retraining., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Knowledge management in optical networks

  • Velasco Esteban, Luis Domingo
  • Tabatabaeimehr, Fatemehsadat
  • Ruiz Ramírez, Marc
©2020 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., Autonomous network operation realized by means of control loops, where prediction from machine learning (ML) models is used as input to proactively reconfigure individual optical devices or the whole optical network, has been recently proposed to minimize human intervention. A general issue in this approach is the limited accuracy of ML models due to the lack of real data for training the models. Although the training dataset can be complemented with data from lab experiments and simulation, it is probable that once in operation, events not considered during the training phase appear thus leading into model inaccuracies. A feasible solution is to implement self-learning approaches, where model inaccuracies are used to re-train the models in the field and to spread such data for training models being used for devices of the same type in other nodes in the network. In this paper, we develop the concept of collective self-learning aiming at improving models error convergence time, as well as at minimizing the amount of data being shared and stored. To this end, we propose a knowledge management (KM) process and an architecture to support it., The research leading to these results has received funding from the European Commission through the METROHAUL project (G.A. nº 761727), from the Spanish MINECO TWINS project (TEC2017-90097-R), and from the Catalan Institution for Research and Advanced Studies (ICREA)., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Experimental demonstration of active and passive optical networks telemetry

  • Gifre Renom, Lluís
  • Izquierdo Zaragoza, José Luis
  • Shariati, Mohammad Behnam
  • Velasco Esteban, Luis Domingo
A distributed architecture enabling active and passive optical network telemetry is presented; a YANG data model is used for remotely configuring monitoring devices for telemetry purposes. Experimental demonstration is carried out for failure localization., Peer Reviewed, Postprint (published version)