Skip to main content

DISEÑANDO UNA INFRAESTRUCTURA DE RED 5G DEFINIDA MEDIANTE CONOCIMIENTO HACIA LA PROXIMA SOCIEDAD DIGITAL

TEC2017-90034-C2-1-R

•
Funding agency name Agencia Estatal de Investigación
Funding agency acronym AEI
Program Programa Estatal de I+D+i Orientado a los Retos de la Sociedad
Call Proyectos I+D+i Retos de Investigación
Call year 2017
State Plan 2017-2020 Plan Estatal de Investigación Científica y Técnica y de Innovación
Beneficiary centre UNIVERSITAT POLITECNICA DE CATALUNYA

Resultados relacionados

Resultados totales (Incluyendo duplicados): 120
Encontrada(s) 2 página(s)

UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

A skewness-aware matrix factorization approach for mesh-structured cloud services

  • Fu, Yongquan
  • Li, Dongsheng
  • Barlet Ros, Pere|||0000-0001-7837-0886
  • Huang, Chun
  • Huang, Zhen
  • Shen, Siqi
  • Su, Huayou
© 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., Online cloud services need to fulfill clients' requests scalably and fast. State-of-the-art cloud services are increasingly deployed as a distributed service mesh. Service to service communication is frequent in the mesh. Unfortunately, problematic events may occur between any pair of nodes in the mesh, therefore, it is vital to maximize the network visibility. A state-of-the-art approach is to model pairwise RTTs based on a latent factor model represented as a low-rank matrix factorization. A latent factor corresponds to a rank-1 component in the factorization model, and is shared by all node pairs. However, different node pairs usually experience a skewed set of hidden factors, which should be fully considered in the model. In this paper, we propose a skewness-aware matrix factorization method named SMF. We decompose the matrix factorization into basic units of rank-one latent factors, and progressively combine rank-one factors for different node pairs. We present a unifying framework to automatically and adaptively select the rank-one factors for each node pair, which not only preserves the low rankness of the matrix model, but also adapts to skewed network latency distributions. Over real-world RTT data sets, SMF significantly improves the relative error by a factor of 0.2 x to 10 x, converges fast and stably, and compactly captures fine-grained local and global network latency structures., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Is machine learning ready for traffic engineering optimization?

  • Bernárdez Gil, Guillermo
  • Suárez-Varela Maciá, José Rafael|||0000-0002-7141-3414
  • López Brescó, Albert|||0009-0005-0611-5516
  • Wu, Bo
  • Xiao, Shihan
  • Cheng, Xiangle
  • Barlet Ros, Pere|||0000-0001-7837-0886
  • Cabellos Aparicio, Alberto|||0000-0001-9329-7584
Traffic Engineering (TE) is a basic building block of the Internet. In this paper, we analyze whether modern Machine Learning (ML) methods are ready to be used for TE optimization. We address this open question through a comparative analysis between the state of the art in ML and the state of the art in TE. To this end, we first present a novel distributed system for TE that leverages the latest advancements in ML. Our system implements a novel architecture that combines Multi-Agent Reinforcement Learning (MARL) and Graph Neural Networks (GNN) to minimize network congestion. In our evaluation, we compare our MARL+GNN system with DEFO, a network optimizer based on Constraint Programming that represents the state of the art in TE. Our experimental results show that the proposed MARL+GNN solution achieves equivalent performance to DEFO in a wide variety of network scenarios including three real-world network topologies. At the same time, we show that MARL+GNN can achieve significant reductions in execution time (from the scale of minutes with DEFO to a few seconds with our solution)., This work was supported by the Spanish MINECO under contract TEC2017-90034-C2-1-R (ALLIANCE), the Catalan Institution for Research and Advanced Studies (ICREA) and the Secretariat for Universities and Research of the Ministry of Business and Knowledge of the Government of Catalonia as well as the European Social Fund., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

NetXplain: Real-time explainability of graph neural networks applied to computer networks

  • Pujol Perich, David
  • Suárez-Varela Maciá, José Rafael|||0000-0002-7141-3414
  • Xiao, Shihan
  • Wu, Bo
  • Cabellos Aparicio, Alberto|||0000-0001-9329-7584
  • Barlet Ros, Pere|||0000-0001-7837-0886
Recent advancements in Deep Learning (DL) have revolutionized the way we can efficiently tackle complex optimization problems. However, existing DL-based solutions are often considered as black boxes due to their high inner complexity. As a result, there is still certain skepticism among the computer network industry about their practical viability to operate data networks. In this context, explainability techniques have recently emerged to unveil why DL models make each decision. This paper focuses on Graph Neural Network (GNN) models applied to computer networks, which have already shown outstanding performance in different network optimization tasks. We thus present NetXplain, a novel real-time explainability solution that uses a GNN to interpret the output produced by another GNN. In the evaluation, we apply the proposed explainability method to RouteNet –a GNN model that predicts end-to-end performance metrics in computer networks. We show that NetXplain operates more than 3 orders of magnitude faster than state-of-the-art explainability solutions when applied to networks up to 24 nodes, which makes this solution compatible with real-time applications. Moreover, it demonstrated strong generalization capabilities over different network scenarios unseen during training., This work has received funding from the European Union’s Horizon 2020 research and innovation programme within the framework of the NGI-POINTER Project funded under grant agreement No. 871528. This paper reflects only the author’s view; the EC is not responsible for any use that may be made of the information it contains. This work was also supported by the Spanish MINECO under contract TEC2017-90034-C2-1-R (ALLIANCE) and the Catalan Institution for Research and Advanced Studies (ICREA)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Joint placement latency optimization of the control plane

  • Rasol, Kurdman Abdulrahman Rasol
  • Domingo Pascual, Jordi|||0000-0001-6277-7542
Software Defined Networking (SDN) is a programmable network architecture that decouples the control plane (controller logic) from the data plane (forwarding plane) and provides flexible network management. A logically centralized control plane supports SDN feasibility. A single controller may not be capable to control the whole network. To tackle this issue, deploying multiple controllers are a necessity to handle a large network. Controller placement problem (CPP) is one of the most important issues in SDN to improve the scalability and influences the latency and performance of the whole network. Most of the previous methods only focused on latency between controllers and their associated switches but ignoring the latency between controllers and the accumulated latency of the control plane.In this paper, we define an accumulated latency to solve CPP, which takes into consideration both the latency between the controller and their associated switches and inter-controller latency. We formulate an optimization problem as a mixed-integer linear programming (MILP) under the constraint of latency. The objective of our study is to minimize the accumulated latency and minimizing the number of network controllers while optimizing their placement for achieving an optimal balance at the same time. The performance of our method is evaluated on the Internet2 OS3E real network topology. Results demonstrate that the proposed method is promising., The work was supported by the Spanish Ministry of Economy and Competitiveness under contract TEC2017-90034-C2-1-R (ALLIANCE)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

Network measurements for web tracking analysis and detection: A tutorial

  • Castell Uroz, Ismael
  • Solé Pareta, Josep|||0000-0002-9411-6308
  • Barlet Ros, Pere|||0000-0001-7837-0886
Digital society has developed to a point where it is nearly impossible for a user to know what it is happening in the background when using the Internet. To understand it, it is necessary to perform network measurements not only at the network layer (e.g., IP, ICMP), but also at the application layer (e.g., HTTP). For example, opening a single website can trigger a cascade of requests to different servers and services to obtain the resources embedded inside it. This process is becoming so complex that, to explore only one website, the number of communications can explode easily from tens to hundreds depending on the website. Inside those communications, there is an ever-increasing portion dedicated to web tracking, a controversial practice from the security and privacy perspective [1]., This work was supported by the Spanish MINECO under contract TEC2017-90034-C2-1-R (ALLIANCE)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Evaluation of Add/Drop flexibility requirements in future SDM-enabled ROADMs

  • Perelló Muntan, Jordi|||0000-0001-6563-2664
  • Gené Bernaus, Joan M.|||0000-0002-1082-9408
  • Klinkowski, Miroslaw
  • Comellas Colomé, Jaume|||0000-0002-9129-0562
  • Spadaro, Salvatore|||0000-0002-4100-1726
In this paper, we evaluate five different architectures for implementing the Add/Drop module of future Spatial Division Multiplexing (SDM)-enabled Re-configurable Optical Add & Drop Multiplexers (ROADMs) with different degrees of flexibility, from a completely static transponder to spatial channel assignment to a fully-flexible transceiver to fibre and spatial channel assignment. The performance of the proposed architectures is compared in a single SDM-enabled ROADM scenario, where local as well as bypass traffic is injected, supported by spectral super-channels at diverse bit-rates and employing different modulation formats. The obtained results illustrate the additional benefits attainable by more flexible Add/Drop architectures versus their increased hardware complexity., The work included in this paper has been supported by the Spanish Government through project ALLIANCE (TEC2017-90034-C2-1-R and TEC2017-90034-C2-2-R) with FEDER contribution. The work of M. Klinkowski was supported by National Science Centre, Poland under Grant 2016/21/B/ST7/02212., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

RouteNet: leveraging graph neural networks for network modeling and optimization in SDN

  • Rusek, Krzysztof
  • Suárez-Varela Maciá, José Rafael|||0000-0002-7141-3414
  • Almasan Puscas, Felician Paul|||0000-0003-3903-6759
  • Barlet Ros, Pere|||0000-0001-7837-0886
  • Cabellos Aparicio, Alberto|||0000-0001-9329-7584
© 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., Network modeling is a key enabler to achieve efficient network operation in future self-driving Software-Defined Networks. However, we still lack functional network models able to produce accurate predictions of Key Performance Indicators (KPI) such as delay, jitter or loss at limited cost. In this paper we propose RouteNet, a novel network model based on Graph Neural Network (GNN) that is able to understand the complex relationship between topology, routing, and input traffic to produce accurate estimates of the per-source/destination per-packet delay distribution and loss. RouteNet leverages the ability of GNNs to learn and model graph-structured information and as a result, our model is able to generalize over arbitrary topologies, routing schemes and traffic intensity. In our evaluation, we show that RouteNet is able to predict accurately the delay distribution (mean delay and jitter) and loss even in topologies, routing and traffic unseen in the training (worst case MRE = 15.4%). Also, we present several use cases where we leverage the KPI predictions of our GNN model to achieve efficient routing optimization and network planning., This work was supported in part by the Polish Ministryof Science and Higher Education with the subvention funds of the Facultyof Computer Science, Electronics and Telecommunications, AGH University,in part by the Spanish MINECO under Contract TEC2017-90034-C2-1-R(ALLIANCE), in part by the Catalan Institution for Research and AdvancedStudies (ICREA) and the FI-AGAUR Grant by the Catalan Government, andin part by PL-Grid Infrastructure., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

Spatial partitioning for proactive spectrum fragmentation avoidance in flex-grid/SDM dynamic optical core networks

  • Comellas Colomé, Jaume|||0000-0002-9129-0562
  • Perelló Muntan, Jordi|||0000-0001-6563-2664
  • Solé Pareta, Josep|||0000-0002-9411-6308
  • Junyent Giralt, Gabriel|||0000-0003-0059-7700
Spectrum fragmentation has always been a major issue to overcome toward spectrally efficient Flex-Grid over Single-Mode Fiber dynamic optical networks and continues like so when evolving to Flex-Grid over Spatial Division Multiplexing (SDM). A possible strategy to eliminate its pernicious effects is to divide the available spectrum into several partitions, dedicating each one of them to only support connections with identical spectral requirements. In this way, a first-fit spectrum assignment ensures that spectral gaps at each spectral partition will always match the bandwidth needs of future connection requests. In this paper, we extend this strategy to be applicable to Flex-Grid/SDM dynamic optical networks. Furthermore, leveraging the spatial multiplicity offered by SDM, we also investigate spatial partitioning as an effective yet simpler and more easily manageable solution to also eliminate the spectrum fragmentation negative effects. Both strategies are numerically evaluated in two reference Flex-Grid/SDM backbone networks with ׿7 spatial multiplicity, yielding noteworthy carried network load gains up to 18% versus a non-partitioned network scenario. When increasing the spatial multiplicities up to ׿30, such gains tend to stabilize around 3–4%. Some results are also obtained under unexpected traffic profile deviations, showing that, even under moderate deviations, partitioning becomes beneficial. Comparing spectral and spatial partitioning, they tend to perform quite similar in all cases. This makes us advocate for spatial partitioning as a more interesting solution for spectrum fragmentation avoidance in this kind of networks., This work has been partially funded by Spanish projects TWINS (TEC2017-90097-R) and ALLIANCE (TEC2017-90034-C2-1-R)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Decision making under uncertainty for the deployment of future networks in IoT scenarios

  • Alzate Mejía, Nestor
  • Santos Boada, Germán|||0000-0002-9528-2287
  • Almeida Amazonas, José Roberto de
The main characteristic of various emerging communication network paradigms in the dimensioning, control and deployment of future networks is the fact that they are human-centric, entailing closelyknit interactions between telematics and human activities. Considering the effect of user behavior, whose dynamics are difficult to model, new uncertainties are introduced in these systems, bringing about network resource management challenges. Within this context, this study seeks to review different decision-making computational methods in conditions of uncertainty for Internet of Things scenarios such as smart spaces, and industry 4.0, through a systematic literature review. According to our research results, a new paradigm for computationally capturing and modeling human behavior context must be developed with the purpose of improving resource management., This work has been partially supported by the Spanish Ministry of Economy and Competitiveness under contract TEC2017-90034-C2-1-R (ALLIANCE project) that receives funding from FEDER. Moreover, it has been partially supported by the Spanish Thematic Network under contract RED2018-102585-T (Go2Edge) and by the aid granted by the Sinfoni project INV2733 of the Cooperative University of Colombia., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

IGNNITION: A framework for fast prototyping of Graph Neural Networks

  • Pujol Perich, David
  • Suárez-Varela Maciá, José Rafael|||0000-0002-7141-3414
  • Ferriol Galmés, Miquel|||0000-0002-7806-2979
  • Xiao, Shihan
  • Wu, Bo
  • Cabellos Aparicio, Alberto|||0000-0001-9329-7584
  • Barlet Ros, Pere|||0000-0001-7837-0886
Recent years have seen the vast potential of Graph Neural Networks (GNN) in many fields where data is structured as graphs (e.g., chemistry, logistics). However, implementing a GNN prototype is still a cumbersome task that requires strong skills in neural network programming. This poses an important barrier to researchers and practitioners that want to apply GNN to their specific problems but do not have the needed Machine Learning expertise. In this paper, we present IGNNITION, a novel open-source framework for fast prototyping of GNNs. This framework is built on top of TensorFlow, and offers an intuitive high-level abstraction that allows the user to define its GNN model via a YAML file, being completely oblivious to the tensor-wise operations made internally by the model. At the same time, IGNNITION offers great flexibility to build any GNN-based architecture. To showcase its versatility, we implement two state-of-the-art GNN models applied to the field of computer networks, which differ considerably from well-known standard GNN architectures. Our evaluation results show that the GNNs produced by IGNNITION are equivalent in performance to implementations directly coded in TensorFlow., This work has received funding from the European Union’s Horizon 2020 research and innovation programme within the framework of the NGI-POINTER Project funded under grant agreement No. 871528. This paper reflects only the authors’ view; the European Commission is not responsible for any use that may be made of the information it contains. This work was also supported by the Spanish MINECO under contract TEC2017-90034-C2-1-R (ALLIANCE) and the Catalan Institution for Research and Advanced Studies (ICREA)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

Routing in optical transport networks with deep reinforcement learning

  • Suárez Varela, José
  • Mestres Sugranyes, Albert
  • Yu, Junlin
  • Kuang, Li
  • Feng, Haoyu
  • Cabellos Aparicio, Alberto|||0000-0001-9329-7584
  • Barlet Ros, Pere|||0000-0001-7837-0886
Deep reinforcement learning (DRL) has recently revolutionized the resolution of decision-making and automated control problems. In the context of networking, there is a growing trend in the research community to apply DRL algorithms to optimization problems such as routing. However, existing proposals fail to achieve good results, often under-performing traditional routing techniques. We argue that the reason behind this poor performance is that they use straightforward representations of networks. In this paper, we propose a DRL-based solution for routing in optical transport networks (OTNs). Contrary to previous works, we propose a more elaborate representation of the network state that reduces the level of knowledge abstraction required for DRL agents and easily captures the singularities of network topologies. Our evaluation results show that using our novel representation, DRL agents achieve better performance and learn how to route traffic in OTNs significantly faster compared to state-of-the-art representations. Additionally, we reverse engineered the routing strategy learned by our DRL agent, and as a result, we found a routing algorithm that outperforms well-known traditional routing heuristics., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

Computing graph neural networks: A survey from algorithms to accelerators

  • Abadal Cavallé, Sergi|||0000-0003-0941-0260
  • Jain, Akshay|||0000-0002-6986-2720
  • Guirado Liñan, Robert
  • López Alonso, Jorge
  • Alarcón Cot, Eduardo José|||0000-0001-7663-7153
Graph Neural Networks (GNNs) have exploded onto the machine learning scene in recent years owing to their capability to model and learn from graph-structured data. Such an ability has strong implications in a wide variety of fields whose data are inherently relational, for which conventional neural networks do not perform well. Indeed, as recent reviews can attest, research in the area of GNNs has grown rapidly and has lead to the development of a variety of GNN algorithm variants as well as to the exploration of ground-breaking applications in chemistry, neurology, electronics, or communication networks, among others. At the current stage research, however, the efficient processing of GNNs is still an open challenge for several reasons. Besides of their novelty, GNNs are hard to compute due to their dependence on the input graph, their combination of dense and very sparse operations, or the need to scale to huge graphs in some applications. In this context, this article aims to make two main contributions. On the one hand, a review of the field of GNNs is presented from the perspective of computing. This includes a brief tutorial on the GNN fundamentals, an overview of the evolution of the field in the last decade, and a summary of operations carried out in the multiple phases of different GNN algorithm variants. On the other hand, an in-depth analysis of current software and hardware acceleration schemes is provided, from which a hardware-software, graph-aware, and communication-centric vision for GNN accelerators is distilled., This work is possible thanks to funding from the European Union’s Horizon 2020 research and innovation programme under Grant No. 863337 (WiPLASH project) and the Spanish Ministry of Economy and Competitiveness under contract TEC2017-90034-C2-1-R (ALLIANCE project) that receives funding from FEDER., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

A hybrid packet/circuit optical transport architecture for DCN

  • Panahi, Nogol
  • Careglio, Davide|||0000-0002-7931-8147
  • Solé Pareta, Josep|||0000-0002-9411-6308
The aim of this paper is to move away from today's multi-tier, manually operated, and performance limited Data Centre Network (DCN) towards more scalable, flexible, and optimized architecture of tomorrow. We propose a new hybrid optical DCN architecture capable of providing a fully optical interconnection path from server to server within and between racks utilizing fast packet switching at the aggregation layer and high-capacity and low latency circuit switching at the core layer. We have built the DCN control plane upon the Software Defined Networks (SDN) paradigm to leverage the added value of emerging optical technologies for providing flexible and programmable heterogeneous IT and network services. Dynamicity, flexibility, and resiliency are combined in the proposed infrastructure to offer advanced services to the data centre providers in order to fulfil the requirements of the upcoming applications., This work has been partially funded by the Spanish Ministry of Economy and Competitiveness under contract FEDER TEC2017-90034-C2-1-R (ALLIANCE project) and supported but not funded by the Generalitat de Catalunya under contract 2017SGR-1037., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Dynamic traffic grooming in Joint Switching (JoS)-enabled flex-grid/SDM optical core networks

  • Rumipamba Zambrano, Rubén Darío
  • Perelló Muntan, Jordi|||0000-0001-6563-2664
  • Spadaro, Salvatore|||0000-0002-4100-1726
We propose and evaluate a dynamic traffic grooming strategy enhancing the performance of JoS-enabled Flex-Grid/SDM networks. Results disclose up to 120% and 44% additional carried network traffic vs. no grooming and existing end-to-end traffic grooming solutions, respectively., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Jellyfish: locality-sensitive subflow sketching

  • Fu, Yongquan
  • An, Lun
  • Shen, Siqi
  • Chen, Kai
  • Barlet Ros, Pere|||0000-0001-7837-0886
To cope with increasing network rates and massive traffic volumes, sketch-based methods have been extensively studied to trade accuracy for memory scalability and storage cost. However, sketches are sensitive to hash collisions due to skewed keys in real world environment, and need complicated performance control for line-rate packet streams.We present Jellyfish, a locality-sensitive sketching framework to address these issues. Jellyfish goes beyond network flow-based sketching towards fragments of network flows called subflows. First, Jellyfish splits consecutive packets from each network flow to subflow records, which not only reduces the rate contention but also provides intermediate subflow representations in form of truncated counters. Next, Jellyfish maps similar subflow records to the same bucket array and merges those from the same network flow to reconstruct the network-flow level counters. Real-world trace-driven experiments show that Jellyfish reduces the average estimation errors by up to six orders of magnitude for per-flow queries, by six orders of magnitude for entropy queries, and up to ten times for heavy-hitter queries., This work was sponsored in part by National Key R&D Program of China under Grant No. 2018YFB0204300, and the National Natural Science Foundation of China (NSFC) under Grant No. 61972409, 61602500, in part by Hong Kong RGC GRF-16215119, and the Spanish MINECO under contract TEC2017-90034-C2-1-R (ALLIANCE)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Unveiling the potential of Graph Neural Networks for network modeling and optimization in SDN

  • Rusek, Krzysztof
  • Suárez-Varela Maciá, José Rafael|||0000-0002-7141-3414
  • Mestres Sugrañes, Albert|||0000-0001-5332-8606
  • Barlet Ros, Pere|||0000-0001-7837-0886
  • Cabellos Aparicio, Alberto|||0000-0001-9329-7584
Network modeling is a critical component for building self-driving Software-Defined Networks, particularly to find optimal routing schemes that meet the goals set by administrators. However, existing modeling techniques do not meet the requirements to provide accurate estimations of relevant performance metrics such as delay and jitter. In this paper we propose a novel Graph Neural Network (GNN) model able to understand the complex relationship between topology, routing and input traffic to produce accurate estimates of the per-source/destination pair mean delay and jitter. GNN are tailored to learn and model information structured as graphs and as a result, our model is able to generalize over arbitrary topologies, routing schemes and variable traffic intensity. In the paper we show that our model provides accurate estimates of delay and jitter (worst case R2 = 0.86) when testing against topologies, routing and traffic not seen during training. In addition, we present the potential of the model for network operation by presenting several use-cases that show its effective use in per-source/destination pair delay/jitter routing optimization and its generalization capabilities by reasoning in topologies and routing schemes not seen during training., This work was supported by AGH University of Science and Technology grant, under contract no. 15.11.230.400, the Spanish MINECO under contract TEC2017-90034-C2-1-R (ALLIANCE) and the Catalan Institution for Research and Advanced Studies (ICREA). The research was also supported in part by PL-Grid Infrastructure., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Multi-level hierarchical controller placement in software defined networking

  • Rasol, Kurdman Abdulrahman Rasol
  • Domingo Pascual, Jordi|||0000-0001-6277-7542
Software Defined Networking (SDN) is a widely used network architecture. It separates the controller logic (or control plane) from forwarding plane (or data plane) to manage the whole network and it enables the network scalability and programmability. One of the most significant challenges in Software Defined Networking (SDN) is the Controller Placement Problem (CPP), which tries to specify the minimum number of controllers and their optimal location. In our study, we extend the methodology based on K-means and K-center algorithms to solve the Controller Placement problem (CPP) into a Multi-level Hierarchical Controller Placement Problem (HCPP), where the Super Controller (SC) is in the top-level, some Master Controllers (MCs) are in the intermediate level and the Domain Controllers (DCs) are at the bottom level. The optimization metric is the latency between controller and switches assigned to it and the latency for controller to controller communication. The proposed architecture and methodology is applied on Western European NRENs topology taken from Internet Topology Zoo. The entire network topology is divided into small scale networks (Clusters) and for each cluster, the optimal number of controllers (Domain Controllers) and their placement is found. A second optimization identifies the optimal number of Master Controllers and their optimal placement. The results validate the methodology and show its feasibility on large networks and different domains. A useful use case may be the deployment of hierarchical levels of controllers for the enforcement of very precise routing policies through different domains., This work was funded by the Spanish Ministry of Economy and Competitiveness under contract TEC2017-90034-C2-1-R (ALLIANCE)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

Random number generation by coherent detection of quantum phase noise

  • Álvarez, Juan Rafael
  • Sarmiento Hernández, Samael|||0000-0001-6549-8426
  • Lázaro Villa, José Antonio|||0000-0002-5592-8434
  • Gené Bernaus, Joan M.|||0000-0002-1082-9408
  • Pérez Torres, Juan|||0000-0002-4454-6676
In 2010 Qi et al. [Opt. Lett. 35(3), 312 (2010)] demonstrated a random number generator based on the drift of the phase of a laser due to spontaneous emission, The out-of-the-lab implementation of this scheme presents two main drawbacks: it requires a long and highly unbalanced interferometer to generate a random phase with uniform probability distribution, or alternatively, a shorter and slightly unbalanced interferometer that notwithstanding requires active stabilization and does not generate a uniform probability distribution without randomness extraction. Here we demonstrate that making use of the random nature of the phase difference between two independent laser sources and two coherent detectors we can overcome these limitations. The two main advantages of the system demonstrated are: i) it generates a probability distribution of quantum origin which is intrinsically uniform and thus in principle needs no randomness extraction for obtaining a uniform distribution, and ii) the phase is measured with telecom equipment routinely used for high capacity coherent optical communications. The speed of random bit generation is determined by the photodetector bandwidth and the linewidth of the lasers. As a by-product of our method, we have obtained images of how phase noise develops with time in a laser. This provides a highly visual alternative way of measuring the coherence time of a laser., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Distributed access control with blockchain

  • Paillissé Vilanova, Jordi|||0000-0001-7733-9713
  • Subirà Nieto, Jordi
  • López Brescó, Albert|||0009-0005-0611-5516
  • Rodríguez Natal, Alberto
  • Ermagan, Vina
  • Maino, Fabio
  • Cabellos Aparicio, Alberto|||0000-0001-9329-7584
The specification and enforcement of network-wide policies in a single administrative domain is common in today's networks and considered as already resolved. However, this is not the case for multi-administrative domains, e.g. among different enterprises. In such situation, new problems arise that challenge classical solutions such as PKIs, which suffer from scalability and granularity concerns. In this paper, we present an extension to Group-Based Policy -a widely used network policy languagefor the aforementioned scenario. To do so, we take advantage of a permissioned blockchain implementation (Hyperledger Fabric) to distribute access control policies in a secure and auditable manner, preserving at the same time the independence of each organization. Network administrators specify polices that are rendered into blockchain transactions. A LISP control plane (RFC 6830) allows routers performing the access control to query the blockchain for authorizations. We have implemented an end-to-end experimental prototype and evaluated it in terms of scalability and network latency., This work has been supported by the Spanish MINECO under contract TEC2017-90034-C2-1-R (ALLIANCE) and the Catalan Institution for Research and Advanced Studies (ICREA)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

Route, modulation format, MIMO and spectrum assignment in Flex-Grid/MCF transparent optical core networks

  • Rumipamba Zambrano, Rubén Darío
  • Perelló Muntan, Jordi|||0000-0001-6563-2664
  • Spadaro, Salvatore|||0000-0002-4100-1726
In this paper, we target an optimal multiple-input multiple-output digital signal processing (MIMO-DSP) assignment to super-channels affected by intercore crosstalk (ICXT) in multicore fiber (MCF) enabled transparent optical core networks. MIMO-DSP undoes ICXT effects, but can be costly with high core density MCFs. Hence, its implementation in the network must be carefully decided. We address our objective as a joint route, modulation format, MIMO and spectrum assignment (RMMSA) problem, for which integer linear programming formulations are provided to optimally solve it in small network scenarios. Moreover, several heuristic approaches are also proposed to solve large-scale problem instances with good accuracy. Their goal is to minimize both network spectral requirements and the amount of MIMO equalized super-channels, taking a crosstalk-free space division multiplexing (SDM) solution as a reference, for example, based on parallel single mode fibers [i.e., a multifiber (MF) scenario]. For our evaluation, we consider several state-of-the-art MCF prototypes and different network topologies. The obtained results, with the considered MCFs, disclose that in national backbone networks, the desirable percentage of super-channels with MIMO equalization to match the performance of an equivalent crosstalk-free SDM solution ranges from 0% to 36, while in continental-wide networks this range raises from 0% to 56%. In addition, in the case of a nonideal MIMO (with a 3 dB/km of crosstalk compensation), such percentages range from 0% to 28% and from 0% to 45% in national and continental-wide backbone networks, respectively, experimenting a performance gap up to 12% with respect to the MF reference scenario., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

NetXplain: Real-time explainability of graph neural networks applied to networking

  • Pujol Perich, David
  • Suárez-Varela Maciá, José Rafael|||0000-0002-7141-3414
  • Xiao, Shihan
  • Wu, Bo
  • Cabellos Aparicio, Alberto|||0000-0001-9329-7584
  • Barlet Ros, Pere|||0000-0001-7837-0886
Recent advancements in Deep Learning (DL) have revolutionized the way we can efficiently tackle complex optimization problems. However, existing DL-based solutions are often considered as black boxes with high inner complexity. As a result, there is still certain skepticism among the networking industry about their practical viability to operate data networks. In this context, explainability techniques have recently emerged to unveil why DL models make each decision. This paper focuses on the explainability of Graph Neural Networks (GNNs) applied to networking. GNNs are a novel DL family with unique properties to generalize over graphs. As a result, they have shown unprecedented performance to solve complex network optimization problems. This paper presents NetXplain, a novel real-time explainability solution that uses a GNN to interpret the output produced by another GNN. In the evaluation, we apply the proposed explainability method to RouteNet, a GNN model that predicts end-to-end QoS metrics in networks. We show that NetXplain operates more than 3 orders of magnitude faster than state-of-the-art explainability solutions when applied to networks up to 24 nodes, which makes it compatible with real-time applications; while demonstrating strong capabilities to generalize to network scenarios not seen during training., This work has been supported by the Spanish MINECO under contract TEC2017‑90034‑C2‑1‑R (ALLIANCE) and the Catalan Institution for Research and Advanced Studies (ICREA)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

SD-access: practical experiences in designing and deploying software defined enterprise networks

  • Paillissé Vilanova, Jordi|||0000-0001-7733-9713
  • Portoles Comeras, Marc
  • López Brescó, Albert|||0009-0005-0611-5516
  • Rodríguez Natal, Alberto
  • Iacobacci, David
  • Leong, Johnson
  • Moreno, Víctor
  • Cabellos Aparicio, Alberto|||0000-0001-9329-7584
  • Maino, Fabio
  • Hooda, Sanjay
Enterprise networks, over the years, have become more and more complex trying to keep up with new requirements that challenge traditional solutions. Just to mention one out of many possible examples, technologies such as Virtual LANs (VLANs) struggle to address the scalability and operational requirements introduced by Internet of Things (IoT) use cases. To keep up with these challenges we have identified four main requirements that are common across modern enterprise networks: (i) scalable mobility, (ii) endpoint segmentation, (iii) simplified administration, and (iv) resource optimization. To address these challenges we designed SDA (Software Defined Access), a solution for modern enterprise networks that leverages Software-Defined Networking (SDN) and other state of the art techniques. In this paper we present the design, implementation and evaluation of SDA. Specifically, SDA: (i) leverages a combination of an overlay approach with an event-driven protocol (LISP) to dynamically adapt to traffic and mobility patterns while preserving resources, and (ii) enforces policies to groups of endpoints for scalable segmentation with low operational burden. We present our experience with deploying SDA in two real-life scenarios: an enterprise campus, and a large warehouse with mobile robots. Our evaluation shows that SDA, when compared with traditional enterprise networks, can (i) reduce overall data plane forwarding state up to 70% thanks to a reactive protocol using a centralized routing server, and (ii) reduce by an order of magnitude the handover delays in scenarios of massive mobility with respect to other approaches. Finally, we discuss lessons learned while deploying and operating SDA, and possible optimizations regarding the use of an event-driven protocol and group-based segmentation., This work was partially supported by the Spanish MINECO under contract TEC2017-90034-C2-1-R (ALLIANCE) and the Catalan Institution for Research and Advanced Studies (ICREA)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

Decision-making under uncertainty for the deployment of future hyperconnected networks: A survey

  • Alzate Mejía, Nestor
  • Santos Boada, Germán|||0000-0002-9528-2287
  • Almeida Amazonas, José Roberto de
This paper is an extended version of our paper published in “Néstor Alzate Mejía, Germán Santos-Boada, José R. de Almeida Amazonas, Decision-Making Under Uncertainty for the Deployment of Future Networks in IoT Scenarios” 11th EAI International Conference on Broadband Communications, Networks, and Systems, 11–12 December 2020., Among the several emerging dimensioning, control and deployment of future communication network paradigms stands out the human-centric characteristic that creates an intricate relationship between telematics and human activities. The hard to model dynamics of user behavior introduces new uncertainties into these systems that give rise to difficult network resource management challenges. According to this context, this work reviews several decision-making computational methods under the influence of uncertainties. This work, by means of a systematic literature review, focuses on sensor-based Internet of Things scenarios such as Smart Spaces and Industry 4.0. According to our conclusions, it is mandatory to establish a means for modeling the human behavior context in order to improve resource assignment and management., This research was funded by Spanish Ministry of Economy and Competitiveness under contract TEC2017-90034-C2-1-R (ALLIANCE project) that received funding from FEDER. Moreover, it was partially supported by the Spanish Thematic Network under contract RED2018-102585-T (Go2Edge) and by the aid granted by the Sinfoni project INV2733 of the Cooperative University of Colombia., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

End-user traffic policing for QoS assurance in polyservice RINA networks

  • León Gaixas, Sergio
  • Perelló Muntan, Jordi|||0000-0001-6563-2664
  • Careglio, Davide|||0000-0002-7931-8147
  • Grasa, Eduard
  • Tarzan Lorente, Miquel
Looking at the ever-increasing amount of heterogeneous distributed applications supported on current data transport networks, it seems evident that best-effort packet delivery falls short to supply their actual needs. Multiple approaches to Quality of Service (QoS) differentiation have been proposed over the years, but their usage has always been hindered by the rigidness of the TCP/IP-based Internet model, which does not even allow for applications to express their QoS needs to the underlying network. In this context, the Recursive InterNetwork Architecture (RINA) has appeared as a clean-slate network architecture aiming to replace the current Internet based on TCP/IP. RINA provides a well-defined QoS support across layers, with standard means for layers to inform of the different QoS guarantees that they can support. Besides, applications and other processes can express their flow requirements, including different QoS-related measures, like delay and jitter, drop probability or average traffic usage. Greedy end-users, however, tend to request the highest quality for their flows, forcing providers to apply intelligent data rate limitation procedures at the edge of their networks. In this work, we propose a new rate limiting policy that, instead of enforcing limits on a per QoS class basis, imposes limits on several independent QoS dimensions. This offers a flexible traffic control to RINA network providers, while enabling end-users freely managing their leased resources. The performance of the proposed policy is assessed in an experimental RINA network test-bed and its performance compared against other policies, either RINA-specific or adopted from TCP/IP. Results show that the proposed policy achieves an effective traffic control for high QoS traffic classes, while also letting lower QoS classes to take profit of the capacity initially reserved for the former ones when available., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

5G connected vehicles supported by optical fiber access

  • Lázaro Villa, José Antonio|||0000-0002-5592-8434
  • Coves Garrido, Massiel
  • Sarmiento Hernández, Samael|||0000-0001-6549-8426
  • Altabas, Jose Antonio
  • Lerin, Adolfo
Future autonomous vehicle will require a high level of computation, communication, sensing and fast response actuation. Communication among vehicles, and vehicles to infrastructure, accessing to remote systems and services will help to reduce other requirements in computation and sensing. This paper revise current state of the art of wireless technologies as: LTE, LTE-A Pro, IEEE802.11p and 5G, as well as optical fiber access technologies as: TDM, WDM, D-WDM and uD-WDM, for the different communications requirements in diverse scenarios as: Vehicle-to-Infrastructure (V2I), Vehicle-to-Vehicle (V2V), Vehicle-to-Pedestrian (V2P) and Vehicle-to-Network (V2N)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Demystifying content-blockers: A large-scale study of actual performance gains

  • Castell Uroz, Ismael
  • Solé Pareta, Josep|||0000-0002-9411-6308
  • Barlet Ros, Pere|||0000-0001-7837-0886
With the evolution of the online advertisement and tracking ecosystem, content-filtering has become the reference tool for improving the security, privacy and browsing experience when surfing the Internet. It is also commonly believed that using content-blockers to stop unsolicited content decreases the time needed for loading websites. In this work, we perform a large-scale study with the 100K most popular websites on the actual performance improvements of using content-blockers. We focus our study on two relevant metrics for measuring the browsing performance; page size and loading time. Our results show that using such tools results in small improvements in terms of page size but, contrary to popular belief, it has a negligible impact in terms of loading time. We also find that, in the case of small and lightweight websites, the use of content-blockers can even result in increased loading times., This work was supported by the Spanish MINECO under contract TEC2017-90034-C2-1-R (ALLIANCE)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

A universal specification for multicore fiber crosstalk

  • Gené Bernaus, Joan M.|||0000-0002-1082-9408
  • Winzer, Peter J.
© 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 order to transition multi-core fiber (MCF) technologies from their research state to volume production, the key design specifications have to be broadly agreed upon, first and foremost an acceptable level of inter-core crosstalk per unit length. Against common belief, we show that MCF crosstalk requirements per unit length are fairly independent of transmission distance in the context of modern coherent optical communication systems. As a consequence, a single value for the tolerable inter-core crosstalk per unit length can be used to specify MCFs, valid from metropolitan (~100 km) to trans-pacific (~10 000 km) deployment scenarios. The notion of a universal inter-core crosstalk specification allows for application-independent MCF designs (including a distance-independent optimum core density) and will facilitate the standardization and volume manufacturing of MCF., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Using spatial division multiplexing to avoid fragmentation in flex-grid optical networks

  • Comellas Colomé, Jaume|||0000-0002-9129-0562
  • Perelló Muntan, Jordi|||0000-0001-6563-2664
  • Solé Pareta, Josep|||0000-0002-9411-6308
  • Junyent Giralt, Gabriel|||0000-0003-0059-7700
Spectrum fragmentation is one of the main drawbacks of Flex-Grid Optical Networks, and it becomes especially harmful under dynamic traffic conditions. When evolving from single fibre networks to spatial division multiplexed (SDM) ones, where many parallel cores/fibres are deployed, spectrum fragmentation still limits their eventual performance. In this work, the total available spectrum is partitioned according to the SDM dimension, and each one of the resulting partitions is assigned to each type of connections. This strategy is proposed as a simple method to eliminate spectrum fragmentation. By allocating all same sized connections onto the same spectral/spatial partition, the randomness in the size of the spectral voids left by torn down connections is eliminated. The proposal is numerically evaluated using different network topologies. The obtained results show that spatial partitioning is an interesting solution for spectrum fragmentation avoidance in these scenarios., This work has been partially funded by Spanish projects TWINS (TEC2017-90097-R) and ALLIANCE (TEC2017-90034-C2-1-R)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Evaluation of joint controller placement for latency and reliability-aware control plane

  • Rasol, Kurdman Abdulrahman Rasol
  • Domingo Pascual, Jordi|||0000-0001-6277-7542
The separation of the forwarding and control planes in software-defined networking provides flexibility for network management. The Controller Placement Problem (CPP) is an important issue affecting network performance. This paper presents an evaluation of the Joint Latency and Reliability-Aware Controller Placement (LRCP) optimization model. LRCP provides network administrators with flexible choices to simultaneously achieve a trade-off between the switch-To-controller latency and the controller-To-controller latency, including the reliability aspect using alternative backup paths. Control plane latency (CPL) is used as the evaluation metric and it is defined as the sum of average switch-To-controller latency and the average inter-controller latency. For each optimal placement in the network, the control plane latency using the real latencies of the real network topology is computed. Results from the control plane latency metric show how the number and location of controllers influence the reliability of the network. In the event of a single link failure, the real CPL for LRCP placements is computed and assesses how good the LRCP placements are. The CPL metric is used to compare with other models using latency and reliability metrics., This publication is part of the Spanish I+D+i project TRAINER-A (ref. PID2020-118011GB-C21), funded by MCIN/AEI/10.13039/501100011033. This work has been also partially funded by the Spanish Ministry of Economy and Competitiveness, under contract TEC 2017-90034-C2-1-R (ALLIANCE)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

Survey on terahertz nanocommunication and networking: A top-down perspective

  • Lemic, Filip
  • Abadal Cavallé, Sergi|||0000-0003-0941-0260
  • Tavernier, Wouter
  • Stroobant, Pieter
  • Colle, Didier
  • Alarcón Cot, Eduardo José|||0000-0001-7663-7153
  • Marquez-Barja, Johann
  • Famaey, Jeroen
Recent developments in nanotechnology herald nanometer-sized devices expected to bring light to a number of groundbreaking applications. Communication with and among nanodevices will be needed for unlocking the full potential of such applications. As the traditional communication approaches cannot be directly applied in nanocommunication, several alternative paradigms have emerged. Among them, electromagnetic nanocommunication in the terahertz (THz) frequency band is particularly promising, mainly due to the breakthrough of novel materials such as graphene. For this reason, numerous research efforts are nowadays targeting THz band nanocommunication and consequently nanonetworking. As it is expected that these trends will continue in the future, we see it beneficial to summarize the current status in these research domains. In this survey, we therefore aim to provide an overview of the current THz nanocommunication and nanonetworking research. Specifically, we discuss the applications envisioned to be supported by nanonetworks operating in the THz band, together with the requirements such applications pose on the underlying nanonetworks. Subsequently, we provide an overview of the current contributions on the different layers of the protocol stack, as well as the available channel models and experimentation tools. Finally, we identify a number of open research challenges and outline several future research directions., The author Filip Lemic was supported by the EU Marie Skłodowska-Curie Actions Individual Fellowships (MSCAIF) project Scalable Localization-enabled In-body Terahertz Nanonetwork (SCaLeITN, grant nr. 893760). The author Pieter Stroobant was supported by a PhD grant of Ghent University’s Special Research Fund (BOF). This work was partly funded by the Ghent University’s BOF/GOA project “Autonomic Networked Multimedia Systems”. This work was also supported by the Spanish Ministry of Economy and Competitiveness under contract TEC2017-90034-C2-1-R (ALLIANCE project) that receives funding from FEDER, and by the European Commission under grants No. 736876 and No. 863337., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Channel characterization for chip-scale wireless communications within computing packages

  • Timoneda i Comas, Xavier
  • Cabellos Aparicio, Alberto|||0000-0001-9329-7584
  • Manessis, Dionyssios
  • Alarcón Cot, Eduardo José|||0000-0001-7663-7153
  • Abadal Cavallé, Sergi|||0000-0003-0941-0260
Wireless Network-on-Chip (WNoC) appears as a promising alternative to conventional interconnect fabrics for chip-scale communications. WNoC takes advantage of an overlaid network composed by a set of millimeter-wave antennas to reduce latency and increase throughput in the communication between cores. Similarly, wireless inter-chip communication has been also proposed to improve the information transfer between processors, memory, and accelerators in multi-chip settings. However, the wireless channel remains largely unknown in both scenarios, especially in the presence of realistic chip packages. This work addresses the issue by accurately modeling flip-chip packages and investigating the propagation both its interior and its surroundings. Through parametric studies, package configurations that minimize path loss are obtained and the trade-offs observed when applying such optimizations are discussed. Single-chip and multi-chip architectures are compared in terms of the path loss exponent, confirming that the amount of bulk silicon found in the pathway between transmitter and receiver is the main determinant of losses., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

Results and achievements of the ALLIANCE Project: New network solutions for 5G and beyond

  • Careglio, Davide|||0000-0002-7931-8147
  • Spadaro, Salvatore|||0000-0002-4100-1726
  • Cabellos Aparicio, Alberto|||0000-0001-9329-7584
  • Lázaro Villa, José Antonio|||0000-0002-5592-8434
  • Barlet Ros, Pere|||0000-0001-7837-0886
  • Gené Bernaus, Joan M.|||0000-0002-1082-9408
  • Perelló Muntan, Jordi|||0000-0001-6563-2664
  • Agraz Bujan, Fernando|||0000-0001-8958-0088
  • Suárez-Varela Maciá, José Rafael|||0000-0002-7141-3414
  • Pagès Raventós, Albert
  • Paillissé Vilanova, Jordi|||0000-0001-7733-9713
  • Almasan Puscas, Felician Paul|||0000-0003-3903-6759
  • Domingo Pascual, Jordi|||0000-0001-6277-7542
  • Solé Pareta, Josep|||0000-0002-9411-6308
Leaving the current 4th generation of mobile communications behind, 5G will represent a disruptive paradigm shift integrating 5G Radio Access Networks (RANs), ultra-high-capacity access/metro/core optical networks, and intra-datacentre (DC) network and computational resources into a single converged 5G network infrastructure. The present paper overviews the main achievements obtained in the ALLIANCE project. This project ambitiously aims at architecting a converged 5G-enabled network infrastructure satisfying those needs to effectively realise the envisioned upcoming Digital Society. In particular, we present two networking solutions for 5G and beyond 5G (B5G), such as Software Defined Networking/Network Function Virtualisation (SDN/NFV) on top of an ultra-high-capacity spatially and spectrally flexible all-optical network infrastructure, and the clean-slate Recursive Inter-Network Architecture (RINA) over packet networks, including access, metro, core and DC segments. The common umbrella of all these solutions is the Knowledge-Defined Networking (KDN)-based orchestration layer which, by implementing Artificial Intelligence (AI) techniques, enables an optimal end-to-end service provisioning. Finally, the cross-layer manager of the ALLIANCE architecture includes two novel elements, namely the monitoring element providing network and user data in real time to the KDN, and the blockchain-based trust element in charge of exchanging reliable and confident information with external domains., This work has been partially funded by the Spanish Ministry of Economy and Competitiveness under contract FEDER TEC2017-90034-C2 (ALLIANCE project) and by the Generalitat de Catalunya under contract 2017SGR-1037 and 2017SGR-605., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Towards real-time routing optimization with deep reinforcement learning: open challenges

  • Almasan Puscas, Felician Paul|||0000-0003-3903-6759
  • Suárez-Varela Maciá, José Rafael|||0000-0002-7141-3414
  • Wu, Bo
  • Xiao, Shihan
  • Barlet Ros, Pere|||0000-0001-7837-0886
  • Cabellos Aparicio, Alberto|||0000-0001-9329-7584
The digital transformation is pushing the existing network technologies towards new horizons, enabling new applications (e.g., vehicular networks). As a result, the networking community has seen a noticeable increase in the requirements of emerging network applications. One main open challenge is the need to accommodate control systems to highly dynamic network scenarios. Nowadays, existing network optimization technologies do not meet the needed requirements to effectively operate in real time. Some of them are based on hand-crafted heuristics with limited performance and adaptability, while some technologies use optimizers which are often too time-consuming. Recent advances in Deep Reinforcement Learning (DRL) have shown a dramatic improvement in decision-making and automated control problems. Consequently, DRL represents a promising technique to efficiently solve a variety of relevant network optimization problems, such as online routing. In this paper, we explore the use of state-of-the-art DRL technologies for real-time routing optimization and outline some relevant open challenges to achieve production-ready DRL-based solutions., This work has received funding from the European Union’s Horizon 2020 research and innovation programme within the framework of the NGI-POINTER Project funded under grant agreement No 871528. This paper reflects only the author’s view; the European Commission is not responsible for any use that may be made of the information it contains. This work was also supported by the Spanish MINECO under contract TEC2017-90034-C2-1-R (ALLIANCE), the Catalan Institution for Research and Advanced Studies (ICREA) and the Secretariat for Universities and Research of the Ministry of Business and Knowledge of the Government of Catalonia and the European Social Fund., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Wide area network autoscaling for cloud applications

  • Serracanta Pujol, Berta|||0000-0003-3195-2576
  • Paillissé Vilanova, Jordi|||0000-0001-7733-9713
  • Cabellos Aparicio, Alberto|||0000-0001-9329-7584
  • Claiborne, Anna
  • Rodríguez Natal, Alberto
  • Ward, Dave
  • Maino, Fabio
Modern cloud orchestrators like Kubernetes provide a versatile and robust way to host applications at scale. One of their key features is autoscaling, which automatically adjusts cloud resources (compute, memory, storage) in order to adapt to the demands of applications. However, the scope of cloud autoscaling is limited to the datacenter hosting the cloud and it doesn't apply uniformly to the allocation of network resources. In I/O-constrained or data-in-motion use cases this can lead to severe performance degradation for the application. For example, when the load on a cloud service increases and the Wide Area Network (WAN) connecting the datacenter to the Internet becomes saturated, the application flows experience an increase in delay and loss. In many cases this is dealt with overprovisioning network capacity, which introduces additional costs and inefficiencies. On the other hand, thanks to the concept of "Network as Code", the WAN exposes a set of APIs that can be used to dynamically allocate and de-allocate capacity on-demand. In this paper we propose extending the concept of cloud autoscaling into the network to address this limitation. This way, applications running in the cloud can communicate their networking requirements, like bandwidth or traffic profile, to a Software-Defined Networking (SDN) controller or Network as a Service (NaaS) platform. Moreover, we aim to define the concepts of vertical and horizontal autoscaling applied to networking. We present a prototype that automatically allocates bandwidth to the underlay network, according to the requirements of the applications hosted in Kubernetes. Finally, we discuss open research challenges., This work was supported by the Spanish MINECO under contract TEC2017-90034-C2-1-R (ALLIANCE), the Catalan Institution for Research and Advanced Studies (ICREA)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

ALLIANCE Project: Architecting a knowledge-defined 5G-enabled network infrastructure

  • Careglio, Davide|||0000-0002-7931-8147
  • Spadaro, Salvatore|||0000-0002-4100-1726
  • Cabellos Aparicio, Alberto|||0000-0001-9329-7584
  • Lázaro Villa, José Antonio|||0000-0002-5592-8434
  • Perelló Muntan, Jordi|||0000-0001-6563-2664
  • Barlet Ros, Pere|||0000-0001-7837-0886
  • Gené Bernaus, Joan M.|||0000-0002-1082-9408
  • Paillissé Vilanova, Jordi|||0000-0001-7733-9713
Leaving the current 4th generation of mobile communications behind, 5G will represent a disruptive paradigm shift integrating access networks, ultra-high capacity access/metro/core optical networks and intra-datacentre network and computational resources into a single converged 5G network infrastructure. Thanks to an extensive deployment of network virtualization techniques leveraged by Software-Defined Networking (SDN) and Network Function Virtualization (NFV) technologies, such a 5G network infrastructure will have to be capable of inter-connecting anything anywhere and over a set of network services truly meeting their diverse communication requirements. In this paper, we overview the ALLIANCE project research activities. ALLIANCE ambitiously aims at architecting a converged 5G-enabled network infrastructure investigating several networking solutions such as SDN/NFV, OpenOverlayRouter (OOR) and the clean-slate Recursive Inter- Network Architecture (RINA) covering access, metro, core and datacentre segments. ALLIANCE relies on a Knowledge-Defined Networking (KDN) orchestrator which take advantage of state-of-the-art machine learning techniques to deploy, operate, monitor and troubleshoot networks automatically., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Experimenting with real application-specific QoS guarantees in a large-scale RINA demonstrator

  • Perelló Muntan, Jordi|||0000-0001-6563-2664
  • López Brescó, Albert|||0009-0005-0611-5516
  • Careglio, Davide|||0000-0002-7931-8147
© 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., This paper reports the definition, setup and obtained results of the Fed4FIRE + medium experiment ERASER, aimed to evaluate the actual Quality of Service (QoS) guarantees that the clean-slate Recursive InterNetwork Architecture (RINA) can deliver to heterogeneous applications at largescale. To this goal, a 37Node 5G metro/regional RINA network scenario, spanning from the enduser to the server where applications run in a datacenter has been configured in the Virtual Wall experimentation facility. This scenario has initially been loaded with synthetic application traffic flows, with diverse QoS requirements, thus reproducing different network load conditions. Next,their experienced QoS metrics endtoend have been measured with two different QTAMux (i.e., the most accepted candidate scheduling policy for providing RINA with its QoS support) deployment scenarios. Moreover, on this RINA network scenario loaded with synthetic application traffic flows, a real HD (1080p) video streaming demonstration has also been conducted, setting up video streaming sessions to endusers at different network locations, illustrating the perceived Quality of Experience (QoE). Obtained results in ERASER disclose that, by appropriately deploying and configuring QTAMux, RINA can yield effective QoS support, which has provided perfect QoE in almost all locations in our demo when assigning video traffic flows the highest (i.e., Gold) QoS Cube., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Towards more realistic network models based on Graph Neural Networks

  • Badia Sampera, Arnau
  • Suárez-Varela Maciá, José Rafael|||0000-0002-7141-3414
  • Almasan Puscas, Felician Paul|||0000-0003-3903-6759
  • Rusek, Krzysztof
  • Barlet Ros, Pere|||0000-0001-7837-0886
  • Cabellos Aparicio, Alberto|||0000-0001-9329-7584
Recently, a Graph Neural Network (GNN) model called RouteNet was proposed as an efficient method to estimate end-to-end network performance metrics such as delay or jitter, given the topology, routing, and traffic of the network. Despite its success in making accurate estimations and generalizing to unseen topologies, the model makes some simplifying assumptions about the network, and does not consider all the particularities of how real networks operate. In this work we extend the architecture of RouteNet to support different features on forwarding devices, specifically we focus on devices with variable queue sizes, and we experimentally evaluate the accuracy of the extended RouteNet architecture., This work was supported by the Spanish MINECO under contract TEC2017-90034-C2-1-R (ALLIANCE), the Catalan Institution for Research and Advanced Studies (ICREA), FI-AGAUR grant by the Catalan Government and the AGH University of Science and by the Polish Ministry of Science and Higher Education with the subvention funds of the Faculty of Computer Science, Electronics and Telecommunications of AGH University. The research was also supported in part by PL-Grid Infrastructure., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Routing based on deep reinforcement learning in optical transport networks

  • Suárez-Varela Maciá, José Rafael|||0000-0002-7141-3414
  • Mestres Sugrañes, Albert|||0000-0001-5332-8606
  • Yu, Junlin
  • Kuang, Li
  • Feng, Haoyu
  • Barlet Ros, Pere|||0000-0001-7837-0886
  • Cabellos Aparicio, Alberto|||0000-0001-9329-7584
This paper addresses the use of Deep Reinforcement Learning for automatic routing in Optical Transport Networks at the electrical-layer level. We propose a DRL-based solution that achieves both high performance and fast learning., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

IGNNITION: fast prototyping of graph neural networks for communication networks

  • Pujol Perich, David
  • Suárez-Varela Maciá, José Rafael|||0000-0002-7141-3414
  • Ferriol Galmés, Miquel|||0000-0002-7806-2979
  • Wu, Bo
  • Xiao, Shihan
  • Cheng, Xiangle
  • Cabellos Aparicio, Alberto|||0000-0001-9329-7584
  • Barlet Ros, Pere|||0000-0001-7837-0886
Graph Neural Networks (GNN) have recently exploded in the Machine Learning area as a novel technique for modeling graph-structured data. This makes them especially suitable for applications in the networking field, as communication networks inherently comprise graphs at many levels (e.g., topology, routing, user connections). In this demo, we will present IGNNITION, an open-source framework for fast prototyping of GNNs applied to communication networks. This framework is especially designed for network engineers and/or researchers with limited background on neural network programming. IGNNITION comprises a set of tools and functionalities that eases and accelerates the whole implementation process, from the design of a GNN model, to its training, evaluation, debugging, and integration into larger network applications. In the demo, we will show how a user can implement a complex GNN model applied to network performance modeling (RouteNet), following three simple steps., This open-source project has received funding from the European Union’s Horizon 2020 research and innovation programme within the framework of the NGI-POINTER Project funded under grant agreement No. 871528. This article reflects only the authors’ view; the EC is not responsible for any use that may be made of the information it contains. The work was also supported by the Spanish MINECO under contract TEC2017-90034-C2-1-R (ALLIANCE) and the Catalan Institution for Research and Advanced Studies (ICREA)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

The graph neural networking challenge: a worldwide competition for education in AI/ML for networks

  • Suárez-Varela Maciá, José Rafael|||0000-0002-7141-3414
  • Ferriol Galmés, Miquel|||0000-0002-7806-2979
  • López Brescó, Albert|||0009-0005-0611-5516
  • Almasan Puscas, Felician Paul|||0000-0003-3903-6759
  • Bernárdez Gil, Guillermo
  • Pujol Perich, David
  • Rusek, Krzysztof
  • Bonniot, Loïck
  • Neumann, Christoph
  • Schnitzler, François
  • Taïani, François
  • Happ, Martin
  • Barlet Ros, Pere|||0000-0001-7837-0886
  • Cabellos Aparicio, Alberto|||0000-0001-9329-7584
During the last decade, Machine Learning (ML) has increasingly become a hot topic in the field of Computer Networks and is expected to be gradually adopted for a plethora of control, monitoring and management tasks in real-world deployments. This poses the need to count on new generations of students, researchers and practitioners with a solid background in ML applied to networks. During 2020, the International Telecommunication Union (ITU) has organized the "ITU AI/ML in 5G challenge", an open global competition that has introduced to a broad audience some of the current main challenges in ML for networks. This large-scale initiative has gathered 23 different challenges proposed by network operators, equipment manufacturers and academia, and has attracted a total of 1300+ participants from 60+ countries. This paper narrates our experience organizing one of the proposed challenges: the "Graph Neural Networking Challenge 2020". We describe the problem presented to participants, the tools and resources provided, some organization aspects and participation statistics, an outline of the top-3 awarded solutions, and a summary with some lessons learned during all this journey. As a result, this challenge leaves a curated set of educational resources openly available to anyone interested in the topic., This work has received funding from the European Union’s H2020 research and innovation programme within the framework of the NGI-POINTER Project funded under grant agreement No. 871528. This paper reflects only the authors’ view; the European Commission is not responsible for any use that may be made of the information it contains. This work was also supported by the Spanish MINECO under contract TEC2017-90034-C2-1-R (ALLIANCE), the Catalan Institution for Research and Advanced Studies (ICREA), and by FI-AGAUR grant by the Catalan Government. Salzburg Research is grateful for the support by the WISS 2025 (Science and Innovation Strategy Salzburg 2025) project ”IDALab Salzburg” (20204-WISS/225/197-2019 and 20102-F1901166-KZP) and the 5G-AI-MLab by the Federal Ministry for Climate Action, Environment, Energy, Mobility, Innovation and Technology (BMK) and the Austrian state Salzburg., Peer Reviewed, Article escrit per 24 autors/autores: José Suárez-Varela (1), Miquel Ferriol-Galmés (1), Albert López (1), Paul Almasan (1), Guillermo Bernárdez (1), David Pujol-Perich (1), Krzysztof Rusek (1,2), Loïck Bonniot (3, 4), Christoph Neumann (3), François Schnitzler (3), François Taïani (4), Martin Happ (5, 6), Christian Maier (5), Jia Lei Du (5), Matthias Herlich (5), Peter Dorfinger (5), Nick Vincent Hainke (7), Stefan Venz (7), Johannes Wegener (7), Henrike Wissing (7), Bo Wu (8), Shihan Xiao (8), Pere Barlet-Ros (1), Albert Cabellos-Aparicio (1). 1- Barcelona Neural Networking center, Universitat Politècnica de Catalunya, Spain. 2- AGH University of Science and Technology, Department of Telecommunications, Poland. 3- InterDigital, France. 4- Univ. Rennes, Inria, CNRS, IRISA, France. 5- Salzburg Research Forschungsgesellschaft mbH, Austria. 6- IDA Lab, University of Salzburg, Austria. 7- Fraunhofer HHI, Germany. 8- Network Technology Lab., Huawei Technologies Co., Ltd., China.
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Detecting cryptocurrency miners with NetFlow/IPFIX network measurements

  • Zayuelas Muñoz, Jordi
  • Suárez-Varela Maciá, José Rafael|||0000-0002-7141-3414
  • Barlet Ros, Pere|||0000-0001-7837-0886
In the last few years, cryptocurrency mining has become more and more important on the Internet activity and nowadays is even having a noticeable impact on the global economy. This has motivated the emergence of a new malicious activity called cryptojacking, which consists of compromising other machines connected to the Internet and leverage their resources to mine cryptocurrencies. In this context, it is of particular interest for network administrators to detect possible cryptocurrency miners using network resources without permission. Currently, it is possible to detect them using IP address lists from known mining pools, processing information from DNS traffic, or directly performing Deep Packet Inspection (DPI) over all the traffic. However, all these methods are still ineffective to detect miners using unknown mining servers or result too expensive to be deployed in real-world networks with large traffic volume. In this paper, we present a machine learning-based method able to detect cryptocurrency miners using NetFlow/IPFIX network measurements. Our method does not require to inspect the packets' payload; as a result, it achieves cost-efficient miner detection with similar accuracy than DPI-based techniques., This work has been supported by the Spanish MINECO under contract TEC2017-90034-C2-1-R (ALLIANCE)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Decentralised Internet infrastructure: Securing inter-domain routing (DEMO)

  • Ferriol Galmés, Miquel|||0000-0002-7806-2979
  • Cabellos Aparicio, Alberto|||0000-0001-9329-7584
The Border Gateway Protocol (BGP) is the inter-domain routing protocol that glues the Internet. BGP does not incorporate security and instead, it relies on careful configuration and manual filtering to offer some protection. As a consequence, the current inter-domain routing infrastructure is partially vulnerable to prefix and path hijacks as well as in misconfigurations that results in route leaks. There are many instances of these vulnerabilities being exploited by malicious actors on the Internet, resulting in disruption of services. To address this issue the IETF has designed RPKI, a centralised trust architecture that relies on Public Key Infrastructure. RPKI has slow adoption and its centralised nature is problematic: network administrators are required to trust CAs and do not have the ultimate control of their own critical Internet resources (e.g,. IP blocks, AS Numbers). In this context, we have built the Decentralised Internet Infrastructure (DII), a distributed ledger to securely store inter-domain routing information. The main advantages of DII are (i) it offers flexible trust models where the Internet community can define the rules of a consensus algorithm that properly reflects the power balance of its members and, (ii) offers protection against vulnerabilities (path hijack and route leaks) that goes well beyond what RPKI offers. We have deployed the prototype on the wild in a worldwide testbed including 7 ASes, we will use the testbed to demonstrate in a realistic scenario how allocation and delegation of Internet resources in DII work, and how this protects ASes against artificially produced path and prefix hijack as well as a route leak., This work was partially supported by the Spanish MINECO under contract TEC2017-90034-C2-1-R (ALLIANCE) and the Catalan Institution for Research and Advanced Studies (ICREA)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

IGNNITION: Bridging the gap between graph neural networks and networking systems

  • Pujol Perich, David
  • Suárez-Varela Maciá, José Rafael|||0000-0002-7141-3414
  • Ferriol Galmés, Miquel|||0000-0002-7806-2979
  • Xiao, Shihan
  • Wu, Bo
  • Cabellos Aparicio, Alberto|||0000-0001-9329-7584
  • Barlet Ros, Pere|||0000-0001-7837-0886
Recent years have seen the vast potential of graph neural networks (GNN) in many fields where data is structured as graphs (e.g., chemistry, recommender systems). In particular, GNNs are becoming increasingly popular in the field of networking, as graphs are intrinsically present at many levels (e.g., topology, routing). The main novelty of GNNs is their ability to generalize to other networks unseen during training, which is an essential feature for developing practical machine learning (ML) solutions for networking. However, implementing a functional GNN prototype is currently a cumbersome task that requires strong skills in neural network programming. This poses an important barrier to network engineers that often do not have the necessary ML expertise. In this article, we present IGNNITION, a novel open source framework that enables fast prototyping of GNNs for networking systems. IGNNITION is based on an intuitive high-level abstraction that hides the complexity behind GNNs, while still offering great flexibility to build custom GNN architectures. To showcase the versatility and performance of this framework, we implement two state-of-the-art GNN models applied to different networking use cases. Our results show that the GNN models produced by IGNNITION are equivalent in terms of accuracy and performance to their native implementations in TensorFlow., This open source project has received funding from the European Union’s Horizon 2020 research and innovation programme within the framework of the NGI-POINTER Project funded under grant agreement No. 871528. This article reflects only the author’s view; the European Commission is not responsible for any use that may be made of the information it contains. The work was also supported by the Spanish MINECO under contract TEC2017-90034-C2-1-R (ALLIANCE) and the Catalan Institution for Research and Advanced Studies (ICREA), Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

Flexible resource provisioning of polarization independent coherent PONs based on non-orthogonal multiple access and multiCAP modulation

  • Izquierdo Nuñez, David
  • Sarmiento Hernández, Samael|||0000-0001-6549-8426
  • Barrio Segura, Miguel
  • Clemente Vicente, Jesús
  • Millán, Pablo
  • Lázaro Villa, José Antonio|||0000-0002-5592-8434
  • Garces Gregorio, Juan Ignacio
  • Altabas Navarro, José Antonio
  • Puerta, Rafael
  • Rommel, Simon
  • Vegas Olmos, Juan J.
  • Tafur Monroy, Idelfonso
A combined non-orthogonal multiple access (NOMA) and multiband carrierless amplitude and phase modulation (multiCAP) scheme is proposed for flexible resource provisioning in coherent passive optical networks (PONs). While the proposed combination increases the data-rate and the number of users through the inclusion of NOMA and multiCAP, the coherent reception increases the range and splitting factor of the network through optical power budget enhancement. The proposed system is experimentally evaluated providing a 20 Gb/s aggregated data-rate per wavelength with 10G optoelectronics and applied in two main PON scenarios with four spectral configurations from full-band CAP to four-band multiCAP. The first PON scenario consists of an existing PON that the network operator requires to increase the number of users; NOMA with full-band CAP or multiCAP can be introduced replacing one or more users of the existing PON by one or several new sub-networks and multiplexing the sub-networks by NOMA. In the other PON scenario, several PONs can be nested sharing an initial splitter and achieving a larger number of users by the combination of NOMA and multiCAP. The proposed techniques are fully compatible with other multiplexing techniques such as dense wavelength division multiplexing, which should be considered to achieve higher numbers of users., Gobierno de Aragón (T20_20R); Ministerio de Economía y Competitividad (TEC2017-85752-R, TEC2017-90034-C2-2-R); Electronic Components and Systems for European Leadership (H2020/ECSEL BRAINE - 876967)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

Immersive interconnected virtual and augmented reality: a 5G and IoT perspective

  • Torres Vega, Maria
  • Liaskos, Christos
  • Abadal Cavallé, Sergi|||0000-0003-0941-0260
  • Papapetrou, Evangelos
  • Jain, Akshay|||0000-0002-6986-2720
  • Mouhouche, Belkacem
  • Kalem, Gokhan
  • Ergüt, Salih
  • Mach, Marian
  • Sabol, Tomas
  • Cabellos Aparicio, Alberto|||0000-0001-9329-7584
  • Grimm, Christoph
  • De Turck, Filip
  • Famaey, Jeroen
The final publication is available at Springer via http://dx.doi.org/10.1007/s10922-020-09545-w, Despite remarkable advances, current augmented and virtual reality (AR/VR) applications are a largely individual and local experience. Interconnected AR/VR, where participants can virtually interact across vast distances, remains a distant dream. The great barrier that stands between current technology and such applications is the stringent end-to-end latency requirement, which should not exceed 20 ms in order to avoid motion sickness and other discomforts. Bringing AR/VR to the next level to enable immersive interconnected AR/VR will require significant advances towards 5G ultra-reliable low-latency communication (URLLC) and a Tactile Internet of Things (IoT). In this article, we articulate the technical challenges to enable a future AR/VR end-to-end architecture, that combines 5G URLLC and Tactile IoT technology to support this next generation of interconnected AR/VR applications. Through the use of IoT sensors and actuators, AR/VR applications will be aware of the environmental and user context, supporting human-centric adaptations of the application logic, and lifelike interactions with the virtual environment. We present potential use cases and the required technological building blocks. For each of them, we delve into the current state of the art and challenges that need to be addressed before the dream of remote AR/VR interaction can become reality., Maria Torres Vega is funded by the Research Foundation Flanders (FWO), grant number 12W4819N. This work has been partially supported by the Spanish Ministry of Economy and Competitiveness under contract TEC2017-90034-C2-1-R (ALLIANCE project) that receives funding from FEDER, Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

On the benefits of probabilistic constellation shaping in Flex-Grid/MCF dynamic optical backbone networks

  • Perelló Muntan, Jordi|||0000-0001-6563-2664
  • Gené Bernaus, Joan M.|||0000-0002-1082-9408
  • Spadaro, Salvatore|||0000-0002-4100-1726
In this paper, we propose a novel methodology for estimating the worst-case spectral efficiency (b/s/Hz) of end-to-end paths across a Multi-Core Fiber (MCF)-enabled Flex-Grid optical network, accounting for physical link and intra-node noise, together with the inter-core crosstalk impairment appearing during the transmission of optical signals over MCFs. In particular, we employ the proposed methodology to quantify the benefits of Probabilistic Constellation Shaping (PCS) versus traditional polarization-multiplexed modulation formats in two reference backbone networks. From the obtained results, PCS yields admissible offered load gains close to 13% under 1% bandwidth blocking probability., This work has been supported by the Spanish Government through project ALLIANCE (TEC2017-90034-C2-1-R and TEC2017-90034-C2-2-R) with FEDER contribution. Moreover, it has been supported by the Spanish Thematic Network under contract RED2018-102585-T (Go2Edge)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

A control plane for WireGuard

  • Paillissé Vilanova, Jordi|||0000-0001-7733-9713
  • Barcia González, Alejandro
  • López Brescó, Albert|||0009-0005-0611-5516
  • Rodríguez Natal, Alberto
  • Maino, Fabio
  • Cabellos Aparicio, Alberto|||0000-0001-9329-7584
WireGuard is a VPN protocol that has gained significant interest recently. Its main advantages are: (i) simple configuration (via pre-shared SSH-like public keys), (ii) mobility support, (iii) reduced codebase to ease auditing, and (iv) Linux kernel implementation that yields high performance. However, WireGuard (intentionally) lacks a control plane. This means that each peer in a WireGuard network has to be manually configured with the other peers’ public key and IP addresses, or by other means. In this paper we present an architecture based on a centralized server to automatically distribute this information. In a nutshell, first we manually establish a WireGuard tunnel to the centralized server, and ask all the peers to store their public keys and IP addresses in it. Then, WireGuard peers use this secure channel to retrieve on-demand the information for the peers they want to communicate to. Our design strives to: (i) offer a key distribution scheme simpler than PKI-based ones, (ii) limit the number of public keys sent to the peers, and (iii) reduce tunnel establishment latency by means of an UDP-based protocol. We argue that such automation can help the deployment in enterprise or ISP scenarios. We also describe in detail our implementation and analyze several performance metrics. Finally, we discuss possible improvements regarding several shortcomings we found during implementation., This work was partially supported by the Spanish MINECO under contract TEC2017-90034-C2-1-R (ALLIANCE) and the Catalan Institution for Research and Advanced Studies (ICREA)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

A multicast routing scheme for the Internet: simulation and experimentation in large-scale networks

  • Careglio, Davide|||0000-0002-7931-8147
  • Agraz Bujan, Fernando|||0000-0001-8958-0088
  • Papadimitriou, Dimitri
With the globalisation of the multimedia entertainment industry and the popularity of streaming and content services, multicast routing is (re-)gaining interest as a bandwidth saving technique. In the 1990’s, multicast routing received a great deal of attention from the research community; nevertheless, its main problems still remain mostly unaddressed and do not reach the acceptance level required for its wide deployment. Among other reasons, the scaling limitation and the relative complexity of the standard multicast protocol architecture can be attributed to the conventional approach of overlaying the multicast routing on top of the unicast routing topology. In this paper, we present the Greedy Compact Multicast Routing (GCMR) scheme. GMCR is characterised by its scalable architecture and independence from any addressing and unicast routing schemes; more specifically, the local knowledge of the cost to direct neighbour nodes is enough for the GCMR scheme to properly operate. The branches of the multicast tree are constructed directly by the joining destination nodes which acquire the routing information needed to reach the multicast source by means of an incremental two-stage search process. In this paper we present the details of GCMR and evaluate its performance in terms of multicast tree size (i.e., the stretch), the memory space consumption, the communication cost, and the transmission cost. The comparative performance analysis is performed against one reference algorithm and two well-known protocol standards. Both simulation and emulation results show that GCMR achieves the expected performance objectives and provide the guidelines for further improvements., This work has been partially funded by the Spanish Ministry of Economy and Competitiveness under contract FEDER TEC2017-90034-C2 (ALLIANCE project) and supported but not funded by the Generalitat de Catalunya under contract 2017SGR-1037., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

TrackSign: guided Web tracking discovery

  • Castell Uroz, Ismael
  • Solé Pareta, Josep|||0000-0002-9411-6308
  • Barlet Ros, Pere|||0000-0001-7837-0886
Current web tracking practices pose a constant threat to the privacy of Internet users. As a result, the research community has recently proposed different tools to combat well-known tracking methods. However, the early detection of new, previously unseen tracking systems is still an open research problem. In this paper, we present TrackSign, a novel approach to discover new web tracking methods. The main idea behind TrackSign is the use of code fingerprinting to identify common pieces of code shared across multiple domains. To detect tracking fingerprints, TrackSign builds a novel 3-mode network graph that captures the relationship between fingerprints, resources and domains. We evaluated TrackSign with the top-100K most popular Internet domains, including almost 1M web resources from more than 5M HTTP requests. Our results show that our method can detect new web tracking resources with high precision (over 92%). TrackSign was able to detect 30K new trackers, more than 10K new tracking resources and 270K new tracking URLs, not yet detected by most popular blacklists. Finally, we also validate the effectiveness of TrackSign with more than 20 years of historical data from the Internet Archive., This work was supported by the Spanish MINECO under contract TEC2017-90034-C2-1-R (ALLIANCE)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

A GRASP meta-heuristic for evaluating the latency and lifetime impact of critical nodes in large wireless sensor networks

  • Sembroiz Ausejo, David
  • Ojaghi Kahjogh, Behnam
  • Careglio, Davide|||0000-0002-7931-8147
  • Ricciardi, Sergio|||0000-0002-0960-421X
Wireless Sensor Networks (WSN) have lately been gaining momentum thanks to the hardware improvements and standardization software efforts. Moreover, the appearance of Internet of Things (IoT) and its reliance on sensors are helping to widely extend the usage of WSNs. However, such networks present drawbacks, mainly because of limited sensor batteries and their vulnerability against physical attacks due to the lack of protection and security. Additionally, not all the sensors inside the network have the same responsibility in terms of traffic handling. In this paper, we firstly analyze the fact that some nodes are more critical than others, considering the most critical node the one that, once incapacitated, causes the most deterioration on the network performance. Such performance is analyzed using two metrics, namely network latency and lifetime. We present a result comparison between a Mixed Integer Programming (MIP) model and a Greedy Randomized Adaptive Search Procedure (GRASP) meta-heuristic for small networks. For bigger networks, GRASP meta-heuristic results are presented to understand how the network degrades as the number of both critical and network nodes increase, by distributing them into two different areas: fixed and incremental to maintain node density., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Preventing route leaks using a decentralized approach: an experimental evaluation

  • Ferriol Galmés, Miquel|||0000-0002-7806-2979
  • Coll Aumatell, Roger
  • Cabellos Aparicio, Alberto|||0000-0001-9329-7584
  • Ren, Shoushou
  • Wei, Xinpeng
  • Liu, Bingyang
In the inter-domain routing infrastructure, a route leak is defined as a violation of the routing policy agreed between two Autonomous Systems (AS). Route leaks have resulted in large-scale outages on the Internet, taking down several services. Although route leaks seem a simple problem, the solution is complex because: (i) ASes consider -partially- routing policy private, (ii) lack of a formal and standard language to express routing policy and (iii) BGP lacks adequate cryptographic-based security. In this paper, we present an experimental analysis of a distributed ledger-based architecture that provides a solution to route leaks. Specifically, the routing policy is unambiguously expressed using a formal language, that is then stored in a blockchain. This decentralized architecture allows private policies and interfaces seamlessly with the current BGP infrastructure, requiring no changes to routers. We build a prototype to evaluate our proposed architecture using Hyperledger, we analyze its performance using a real-world BGP dataset. Our results show that our architecture scales linearly with relevant metrics. Additionally, we validate the architecture preventing an artificially introduced route leak in a realistic 10 AS topology., This work was supported by the Spanish MINECO under contract TEC2017-90034-C2-1-R (ALLIANCE) and the Catalan Institution for Research and Advanced Studies (ICREA)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

Greenfield gradual migration planning toward spectrally-spatially flexible optical networks

  • Lechowicz, Piotr
  • Goscien, Roza
  • Rumipamba Zambrano, Rubén Darío
  • Perelló Muntan, Jordi|||0000-0001-6563-2664
  • Spadaro, Salvatore|||0000-0002-4100-1726
  • Walkowiak, Krzysztof
© 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, This article identifies the unavoidably required upgrade of short-term realizable elastic optical networks (EONs) operating over single-mode fibers toward spectrally-spatially flexible optical networks (SS-FONs), exploiting spatial division multiplexing (SDM) technology in order to cope with the expected mid- and long-term future traffic forecasts. Since a complete EON to SS-FON network migration overnight is unrealistic, this article proposes a gradual greenfield migration strategy employing novel heuristic planning solutions. These solutions allow identifying which network components should be SDM-capable so as to support the forecasted traffic increase at a given time. To allow transmission in an SS-FON, links should be extended with SDM-capable fibers and nodes should be equipped with SDM-capable reconfigurable optical add/drop multiplexers. Using the proposed solutions, a migration case study in a national reference core network is provided and analyzed., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

Decentralized trust in the inter-domain routing infrastructure

  • Paillissé Vilanova, Jordi|||0000-0001-7733-9713
  • Manrique, Jan
  • Bonet, Guillem
  • Rodríguez Natal, Alberto
  • Maino, Fabio
  • Cabellos Aparicio, Alberto|||0000-0001-9329-7584
Inter-domain routing security is of critical importance to the Internet since it prevents unwanted traffic redirections. The current system is based on a Public Key Infrastructure (PKI), a centralized repository of digital certificates. However, the inherent centralization of such design creates tensions between its participants and hinders its deployment. In addition, some technical drawbacks of PKIs delay widespread adoption. In this paper we present IPchain, a blockchain to store the allocations and delegations of IP addresses. IPchain leverages blockchains' properties to decentralize trust among its participants, with the final goal of providing flexible trust models that adapt better to the ever-changing geopolitical landscape. Moreover, we argue that Proof of Stake is a suitable consensus algorithm for IPchain due to the unique incentive structure of this use-case, and that blockchains offer relevant technical advantages when compared to existing systems, such as simplified management. In order to show its feasibility and suitability, we have implemented and evaluated IPchain's performance and scalability storing around 350k IP prefixes in a 2.5 GB chain., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

IPchain: securing IP prefix allocation and delegation with blockchain

  • Paillissé Vilanova, Jordi|||0000-0001-7733-9713
  • Ferriol Galmés, Miquel|||0000-0002-7806-2979
  • Garcia Ribera, Eric
  • Latif Martínez, Hamid|||0000-0001-6006-3175
  • Piris Yamuza, Carlos
  • López Brescó, Albert|||0009-0005-0611-5516
  • Cabellos Aparicio, Alberto|||0000-0001-9329-7584
We present IPchain, a blockchain to store the allocations and delegations of IP addresses, with the aim of easing the deployment of secure interdomain routing systems. Interdomain routing security is of vital importance to the Internet since it prevents unwanted traffic redirections. IPchain makes use of blockchains' properties to provide flexible trust models and simplified management when compared to existing systems. In this paper we argue that Proof of Stake is a suitable consensus algorithm for IPchain due to the unique incentive structure of this use-case. We have implemented and evaluated IPchain's performance and scalability storing around 150k IP prefixes in a 1GB chain., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Towards accurate classification of HTTPS traffic in Software-Defined Networks

  • Suárez-Varela Maciá, José Rafael|||0000-0002-7141-3414
  • Barlet Ros, Pere|||0000-0001-7837-0886
In nowadays Internet, there is a strong trend to encrypt the traffic in order to protect users' privacy. This results in a hard challenge for traffic classification, as the payload in the packets cannot be accessed anymore. In this context, some techniques were proposed for traditional networks in order to classify this traffic. However, we could not find previous works addressing encrypted traffic classification considering the particularities of the Software-Defined Networking (SDN) paradigm. In this paper we present an OpenFlow-based classification system which combines techniques leveraging information in SSL/TLS certificates and DNS traffic to perform accurate flow-level classification for encrypted traffic. We make experiments with real-world traffic to evaluate the overall classification accuracy of our system as well as the accuracy detecting specific popular applications (e.g., Netflix). Furthermore, we assess the processing overhead when deploying our system in SDN environments. As a result, we observe that the support provided by OpenFlow enables to achieve a high accuracy with a more reduced processing overhead than in traditional networks, where typically the whole traffic is mirrored to an external collector that classifies the traffic., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Architecting a secure wireless network-on-chip

  • Lebiednik, Brian
  • Abadal Cavallé, Sergi|||0000-0003-0941-0260
  • Kwon, Hyoukjun
  • Krishna, Tushar
With increasing integration in SoCs, the Network-on-Chip (NoC) connecting cores and accelerators is of paramount importance to provide low-latency and high-throughput communication. Due to limits to scaling of electrical wires in terms of energy and delay, especially for long multi-mm distances on-chip, alternate technologies such as Wireless Network-on-Chip (WNoC) have shown promise. WNoCs can provide low-latency one-hop broadcasts across the entire chip and can augment point-to-point multi-hop signaling over traditional wired NoCs. Thus, there has been a recent surge in research demonstrating the performance and energy benefits of WNoCs. However, little to no work has studied the additional security and fault tolerance challenges that are unique to WNoCs. In this work, we study potential threats related to denial-of-service, spoofing, and eavesdropping attacks in WNoCs, due to malicious hardware trojans or faulty wireless components. We introduce Prometheus, a dropin solution inside the network interface that provides protection from all three attacks, while adhering to the strict area, power and latency constraints of on-chip systems., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Feature engineering for deep reinforcement learning based routing

  • Suárez-Varela Maciá, José Rafael|||0000-0002-7141-3414
  • Mestres Sugrañes, Albert|||0000-0001-5332-8606
  • Yu, Junlin
  • Kuang, Li
  • Feng, Haoyu
  • Barlet Ros, Pere|||0000-0001-7837-0886
  • Cabellos Aparicio, Alberto|||0000-0001-9329-7584
Recent advances in Deep Reinforcement Learning (DRL) techniques are providing a dramatic improvement in decision-making and automated control problems. As a result, we are witnessing a growing number of research works that are proposing ways of applying DRL techniques to network-related problems such as routing. However, such proposals failed to achieve good results, often under-performing traditional routing techniques. We argue that successfully applying DRL-based techniques to networking requires finding good representations of the network parameters: feature engineering. DRL agents need to represent both the state (e.g., link utilization) and the action space (e.g., changes to the routing policy). In this paper, we show that existing approaches use straightforward representations that lead to poor performance. We propose a novel representation of the state and action that outperforms existing ones and that is flexible enough to be applied to many networking use-cases. We test our representation in two different scenarios: (i) routing in optical transport networks and (ii) QoS-aware routing in IP networks. Our results show that the DRL agent achieves significantly better performance compared to existing state/action representations., This work has been supported by the Spanish MINECO under contract TEC2017-90034-C2-1-R (ALLIANCE) and the Catalan Institution for Research and Advanced Studies (ICREA)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Challenging the generalization capabilities of Graph Neural Networks for network modeling

  • Suárez-Varela Maciá, José Rafael|||0000-0002-7141-3414
  • Carol Bosch, Sergi
  • Rusek, Krzysztof
  • Almasan Puscas, Felician Paul|||0000-0003-3903-6759
  • Arias Vicente, Marta|||0000-0001-7359-1815
  • Barlet Ros, Pere|||0000-0001-7837-0886
  • Cabellos Aparicio, Alberto|||0000-0001-9329-7584
Today, network operators still lack functional network models able to make accurate predictions of end-to-end Key Performance Indicators (e.g., delay or jitter) at limited cost. Recently a novel Graph Neural Network (GNN) model called RouteNet was proposed as a cost-effective alternative to estimate the per-source/destination pair mean delay and jitter in networks. Thanks to its GNN architecture that operates over graph-structured data, RouteNet revealed an unprecedented ability to learn and model the complex relationships among topology, routing and input traffic in networks. As a result, it was able to make performance predictions with similar accuracy than resource-hungry packet-level simulators even in network scenarios unseen during training. In this demo, we will challenge the generalization capabilities of RouteNet with more complex scenarios, including larger topologies., This work was supported by the Spanish MINECO under contract TEC2017-90034-C2-1-R (ALLIANCE), the Catalan Institution for Research and Advanced Studies (ICREA) and the AGH University of Science and Technology grant, under contract no. 15.11.230.400. The research was also supported in part by PL-Grid Infrastructure., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Joint latency and reliability-aware controller placement

  • Rasol, Kurdman Abdulrahman Rasol
  • Domingo Pascual, Jordi|||0000-0001-6277-7542
In network architectures based on Software Defined Networking (SDN) the control plane (control logic) is separated from the network data plane (forwarding plane) while traditional network routers combine both. Software Defined networks facilitates a centralized networking system where a logical controller manages the global view of the network. In this paper, we first propose a new metric on the controller placement problem (CPP) that simultaneously considers the communication latency and communication reliability both between switches and controllers and between controllers. Reliability is considered for single-link failure. We model the problem of determining the optimal controller placement to provide low latencies in the control plane traffic. The objective of this study is to minimize the average accumulated latency by jointly taking into account the latency between controller to switches and inter-controller while optimizing their placement for achieving an optimal balance simultaneously. The optimization problem is formulated as a mixed-integer linear programming (MILP) model under the constraints of latency and reliability. We evaluated the performance of our proposed metric by using the Internet2 OS3E network topology. Different from previous work, we focus on the control traffic exchanged among controllers to synchronize their shared data structure. Results demonstrate that the proposed method is promising., This research was supported by the Spanish Ministry of Economy and Competitiveness under contract TEC2017-90034-C2-1-R (ALLIANCE)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

A machine learning-based approach for virtual network function modeling

  • Mestres Sugrañes, Albert|||0000-0001-5332-8606
  • Alarcón Cot, Eduardo José|||0000-0001-7663-7153
  • Cabellos Aparicio, Alberto|||0000-0001-9329-7584
Recent trends in networking are proposing the use of Machine Learning (ML) techniques for the control and operation of the network. The application of ML to networking brings several use-cases as well as challenges. The objective of this paper is to explore the feasibility of applying different models and ML techniques to model complex networks elements, such as Virtual Network Functions (VNFs). In particular, we focus on the characterization of the CPU consumption of the VNF as a function of the characteristics of the input traffic. The traffic is represented by a set of features describing characteristics from the transport layer to the application layer in small time batches. The CPU consumption is observed from the hypervisor and corresponds to the average CPU consumption when the traffic batch is processed. We experimentally demonstrate that we can learn the behavior of different VNF in order to model its CPU consumption. We conclude that the behavior of different VNF can be modeled using ML techniques. © 2018 IEEE., Peer Reviewed
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

A GRASP meta-heuristic for evaluating the latency and lifetime impact of critical nodes in large wireless sensor networks

  • Sembroiz Ausejo, David
  • Ojaghi Kahjogh, Behnam
  • Careglio, Davide
  • Ricciardi, Sergio
Wireless Sensor Networks (WSN) have lately been gaining momentum thanks to the hardware improvements and standardization software efforts. Moreover, the appearance of Internet of Things (IoT) and its reliance on sensors are helping to widely extend the usage of WSNs. However, such networks present drawbacks, mainly because of limited sensor batteries and their vulnerability against physical attacks due to the lack of protection and security. Additionally, not all the sensors inside the network have the same responsibility in terms of traffic handling. In this paper, we firstly analyze the fact that some nodes are more critical than others, considering the most critical node the one that, once incapacitated, causes the most deterioration on the network performance. Such performance is analyzed using two metrics, namely network latency and lifetime. We present a result comparison between a Mixed Integer Programming (MIP) model and a Greedy Randomized Adaptive Search Procedure (GRASP) meta-heuristic for small networks. For bigger networks, GRASP meta-heuristic results are presented to understand how the network degrades as the number of both critical and network nodes increase, by distributing them into two different areas: fixed and incremental to maintain node density., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Feature engineering for deep reinforcement learning based routing

  • Suárez-Varela Maciá, José Rafael
  • Mestres Sugrañes, Albert
  • Yu, Junlin
  • Kuang, Li
  • Feng, Haoyu
  • Barlet Ros, Pere
  • Cabellos Aparicio, Alberto
Recent advances in Deep Reinforcement Learning (DRL) techniques are providing a dramatic improvement in decision-making and automated control problems. As a result, we are witnessing a growing number of research works that are proposing ways of applying DRL techniques to network-related problems such as routing. However, such proposals failed to achieve good results, often under-performing traditional routing techniques. We argue that successfully applying DRL-based techniques to networking requires finding good representations of the network parameters: feature engineering. DRL agents need to represent both the state (e.g., link utilization) and the action space (e.g., changes to the routing policy). In this paper, we show that existing approaches use straightforward representations that lead to poor performance. We propose a novel representation of the state and action that outperforms existing ones and that is flexible enough to be applied to many networking use-cases. We test our representation in two different scenarios: (i) routing in optical transport networks and (ii) QoS-aware routing in IP networks. Our results show that the DRL agent achieves significantly better performance compared to existing state/action representations., This work has been supported by the Spanish MINECO under contract TEC2017-90034-C2-1-R (ALLIANCE) and the Catalan Institution for Research and Advanced Studies (ICREA)., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

End-user traffic policing for QoS assurance in polyservice RINA networks

  • León Gaixas, Sergio
  • Perelló Muntan, Jordi
  • Careglio, Davide
  • Grasa, Eduard
  • Tarzan Lorente, Miquel
Looking at the ever-increasing amount of heterogeneous distributed applications supported on current data transport networks, it seems evident that best-effort packet delivery falls short to supply their actual needs. Multiple approaches to Quality of Service (QoS) differentiation have been proposed over the years, but their usage has always been hindered by the rigidness of the TCP/IP-based Internet model, which does not even allow for applications to express their QoS needs to the underlying network. In this context, the Recursive InterNetwork Architecture (RINA) has appeared as a clean-slate network architecture aiming to replace the current Internet based on TCP/IP. RINA provides a well-defined QoS support across layers, with standard means for layers to inform of the different QoS guarantees that they can support. Besides, applications and other processes can express their flow requirements, including different QoS-related measures, like delay and jitter, drop probability or average traffic usage. Greedy end-users, however, tend to request the highest quality for their flows, forcing providers to apply intelligent data rate limitation procedures at the edge of their networks. In this work, we propose a new rate limiting policy that, instead of enforcing limits on a per QoS class basis, imposes limits on several independent QoS dimensions. This offers a flexible traffic control to RINA network providers, while enabling end-users freely managing their leased resources. The performance of the proposed policy is assessed in an experimental RINA network test-bed and its performance compared against other policies, either RINA-specific or adopted from TCP/IP. Results show that the proposed policy achieves an effective traffic control for high QoS traffic classes, while also letting lower QoS classes to take profit of the capacity initially reserved for the former ones when available., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

NetXplain: Real-time explainability of graph neural networks applied to computer networks

  • Pujol Perich, David
  • Suárez-Varela Maciá, José Rafael
  • Xiao, Shihan
  • Wu, Bo
  • Cabellos Aparicio, Alberto
  • Barlet Ros, Pere
Recent advancements in Deep Learning (DL) have revolutionized the way we can efficiently tackle complex optimization problems. However, existing DL-based solutions are often considered as black boxes due to their high inner complexity. As a result, there is still certain skepticism among the computer network industry about their practical viability to operate data networks. In this context, explainability techniques have recently emerged to unveil why DL models make each decision. This paper focuses on Graph Neural Network (GNN) models applied to computer networks, which have already shown outstanding performance in different network optimization tasks. We thus present NetXplain, a novel real-time explainability solution that uses a GNN to interpret the output produced by another GNN. In the evaluation, we apply the proposed explainability method to RouteNet –a GNN model that predicts end-to-end performance metrics in computer networks. We show that NetXplain operates more than 3 orders of magnitude faster than state-of-the-art explainability solutions when applied to networks up to 24 nodes, which makes this solution compatible with real-time applications. Moreover, it demonstrated strong generalization capabilities over different network scenarios unseen during training., This work has received funding from the European Union’s Horizon 2020 research and innovation programme within the framework of the NGI-POINTER Project funded under grant agreement No. 871528. This paper reflects only the author’s view; the EC is not responsible for any use that may be made of the information it contains. This work was also supported by the Spanish MINECO under contract TEC2017-90034-C2-1-R (ALLIANCE) and 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)

IPchain: securing IP prefix allocation and delegation with blockchain

  • Paillissé Vilanova, Jordi
  • Ferriol Galmés, Miquel
  • Garcia Ribera, Eric
  • Latif Martínez, Hamid
  • Piris Yamuza, Carlos
  • López Brescó, Albert
  • Cabellos Aparicio, Alberto
We present IPchain, a blockchain to store the allocations and delegations of IP addresses, with the aim of easing the deployment of secure interdomain routing systems. Interdomain routing security is of vital importance to the Internet since it prevents unwanted traffic redirections. IPchain makes use of blockchains' properties to provide flexible trust models and simplified management when compared to existing systems. In this paper we argue that Proof of Stake is a suitable consensus algorithm for IPchain due to the unique incentive structure of this use-case. We have implemented and evaluated IPchain's performance and scalability storing around 150k IP prefixes in a 1GB chain., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Dynamic traffic grooming in Joint Switching (JoS)-enabled flex-grid/SDM optical core networks

  • Rumipamba Zambrano, Rubén Darío
  • Perelló Muntan, Jordi
  • Spadaro, Salvatore
We propose and evaluate a dynamic traffic grooming strategy enhancing the performance of JoS-enabled Flex-Grid/SDM networks. Results disclose up to 120% and 44% additional carried network traffic vs. no grooming and existing end-to-end traffic grooming solutions, respectively., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

A multicast routing scheme for the Internet: simulation and experimentation in large-scale networks

  • Careglio, Davide
  • Agraz Bujan, Fernando
  • Papadimitriou, Dimitri
With the globalisation of the multimedia entertainment industry and the popularity of streaming and content services, multicast routing is (re-)gaining interest as a bandwidth saving technique. In the 1990’s, multicast routing received a great deal of attention from the research community; nevertheless, its main problems still remain mostly unaddressed and do not reach the acceptance level required for its wide deployment. Among other reasons, the scaling limitation and the relative complexity of the standard multicast protocol architecture can be attributed to the conventional approach of overlaying the multicast routing on top of the unicast routing topology. In this paper, we present the Greedy Compact Multicast Routing (GCMR) scheme. GMCR is characterised by its scalable architecture and independence from any addressing and unicast routing schemes; more specifically, the local knowledge of the cost to direct neighbour nodes is enough for the GCMR scheme to properly operate. The branches of the multicast tree are constructed directly by the joining destination nodes which acquire the routing information needed to reach the multicast source by means of an incremental two-stage search process. In this paper we present the details of GCMR and evaluate its performance in terms of multicast tree size (i.e., the stretch), the memory space consumption, the communication cost, and the transmission cost. The comparative performance analysis is performed against one reference algorithm and two well-known protocol standards. Both simulation and emulation results show that GCMR achieves the expected performance objectives and provide the guidelines for further improvements., This work has been partially funded by the Spanish Ministry of Economy and Competitiveness under contract FEDER TEC2017-90034-C2 (ALLIANCE project) and supported but not funded by the Generalitat de Catalunya under contract 2017SGR-1037., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Routing based on deep reinforcement learning in optical transport networks

  • Suárez-Varela Maciá, José Rafael
  • Mestres Sugrañes, Albert
  • Yu, Junlin
  • Kuang, Li
  • Feng, Haoyu
  • Barlet Ros, Pere
  • Cabellos Aparicio, Alberto
This paper addresses the use of Deep Reinforcement Learning for automatic routing in Optical Transport Networks at the electrical-layer level. We propose a DRL-based solution that achieves both high performance and fast learning., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Evaluation of joint controller placement for latency and reliability-aware control plane

  • Rasol, Kurdman Abdulrahman Rasol
  • Domingo Pascual, Jordi
The separation of the forwarding and control planes in software-defined networking provides flexibility for network management. The Controller Placement Problem (CPP) is an important issue affecting network performance. This paper presents an evaluation of the Joint Latency and Reliability-Aware Controller Placement (LRCP) optimization model. LRCP provides network administrators with flexible choices to simultaneously achieve a trade-off between the switch-To-controller latency and the controller-To-controller latency, including the reliability aspect using alternative backup paths. Control plane latency (CPL) is used as the evaluation metric and it is defined as the sum of average switch-To-controller latency and the average inter-controller latency. For each optimal placement in the network, the control plane latency using the real latencies of the real network topology is computed. Results from the control plane latency metric show how the number and location of controllers influence the reliability of the network. In the event of a single link failure, the real CPL for LRCP placements is computed and assesses how good the LRCP placements are. The CPL metric is used to compare with other models using latency and reliability metrics., This publication is part of the Spanish I+D+i project TRAINER-A (ref. PID2020-118011GB-C21), funded by MCIN/AEI/10.13039/501100011033. This work has been also partially funded by the Spanish Ministry of Economy and Competitiveness, under contract TEC 2017-90034-C2-1-R (ALLIANCE)., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

IGNNITION: fast prototyping of graph neural networks for communication networks

  • Pujol Perich, David
  • Suárez-Varela Maciá, José Rafael
  • Ferriol Galmés, Miquel
  • Wu, Bo
  • Xiao, Shihan
  • Cheng, Xiangle
  • Cabellos Aparicio, Alberto
  • Barlet Ros, Pere
Graph Neural Networks (GNN) have recently exploded in the Machine Learning area as a novel technique for modeling graph-structured data. This makes them especially suitable for applications in the networking field, as communication networks inherently comprise graphs at many levels (e.g., topology, routing, user connections). In this demo, we will present IGNNITION, an open-source framework for fast prototyping of GNNs applied to communication networks. This framework is especially designed for network engineers and/or researchers with limited background on neural network programming. IGNNITION comprises a set of tools and functionalities that eases and accelerates the whole implementation process, from the design of a GNN model, to its training, evaluation, debugging, and integration into larger network applications. In the demo, we will show how a user can implement a complex GNN model applied to network performance modeling (RouteNet), following three simple steps., This open-source project has received funding from the European Union’s Horizon 2020 research and innovation programme within the framework of the NGI-POINTER Project funded under grant agreement No. 871528. This article reflects only the authors’ view; the EC is not responsible for any use that may be made of the information it contains. The work was also supported by the Spanish MINECO under contract TEC2017-90034-C2-1-R (ALLIANCE) and the Catalan Institution for Research and Advanced Studies (ICREA)., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Survey on terahertz nanocommunication and networking: A top-down perspective

  • Lemic, Filip
  • Abadal Cavallé, Sergi
  • Tavernier, Wouter
  • Stroobant, Pieter
  • Colle, Didier
  • Alarcón Cot, Eduardo José
  • Marquez-Barja, Johann
  • Famaey, Jeroen
Recent developments in nanotechnology herald nanometer-sized devices expected to bring light to a number of groundbreaking applications. Communication with and among nanodevices will be needed for unlocking the full potential of such applications. As the traditional communication approaches cannot be directly applied in nanocommunication, several alternative paradigms have emerged. Among them, electromagnetic nanocommunication in the terahertz (THz) frequency band is particularly promising, mainly due to the breakthrough of novel materials such as graphene. For this reason, numerous research efforts are nowadays targeting THz band nanocommunication and consequently nanonetworking. As it is expected that these trends will continue in the future, we see it beneficial to summarize the current status in these research domains. In this survey, we therefore aim to provide an overview of the current THz nanocommunication and nanonetworking research. Specifically, we discuss the applications envisioned to be supported by nanonetworks operating in the THz band, together with the requirements such applications pose on the underlying nanonetworks. Subsequently, we provide an overview of the current contributions on the different layers of the protocol stack, as well as the available channel models and experimentation tools. Finally, we identify a number of open research challenges and outline several future research directions., The author Filip Lemic was supported by the EU Marie Skłodowska-Curie Actions Individual Fellowships (MSCAIF) project Scalable Localization-enabled In-body Terahertz Nanonetwork (SCaLeITN, grant nr. 893760). The author Pieter Stroobant was supported by a PhD grant of Ghent University’s Special Research Fund (BOF). This work was partly funded by the Ghent University’s BOF/GOA project “Autonomic Networked Multimedia Systems”. This work was also supported by the Spanish Ministry of Economy and Competitiveness under contract TEC2017-90034-C2-1-R (ALLIANCE project) that receives funding from FEDER, and by the European Commission under grants No. 736876 and No. 863337., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

A universal specification for multicore fiber crosstalk

  • Gené Bernaus, Joan M.
  • Winzer, Peter J.
© 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 order to transition multi-core fiber (MCF) technologies from their research state to volume production, the key design specifications have to be broadly agreed upon, first and foremost an acceptable level of inter-core crosstalk per unit length. Against common belief, we show that MCF crosstalk requirements per unit length are fairly independent of transmission distance in the context of modern coherent optical communication systems. As a consequence, a single value for the tolerable inter-core crosstalk per unit length can be used to specify MCFs, valid from metropolitan (~100 km) to trans-pacific (~10 000 km) deployment scenarios. The notion of a universal inter-core crosstalk specification allows for application-independent MCF designs (including a distance-independent optimum core density) and will facilitate the standardization and volume manufacturing of MCF., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Demystifying content-blockers: A large-scale study of actual performance gains

  • Castell Uroz, Ismael
  • Solé Pareta, Josep
  • Barlet Ros, Pere
With the evolution of the online advertisement and tracking ecosystem, content-filtering has become the reference tool for improving the security, privacy and browsing experience when surfing the Internet. It is also commonly believed that using content-blockers to stop unsolicited content decreases the time needed for loading websites. In this work, we perform a large-scale study with the 100K most popular websites on the actual performance improvements of using content-blockers. We focus our study on two relevant metrics for measuring the browsing performance; page size and loading time. Our results show that using such tools results in small improvements in terms of page size but, contrary to popular belief, it has a negligible impact in terms of loading time. We also find that, in the case of small and lightweight websites, the use of content-blockers can even result in increased loading times., This work was supported by the Spanish MINECO under contract TEC2017-90034-C2-1-R (ALLIANCE)., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Decision-making under uncertainty for the deployment of future hyperconnected networks: A survey

  • Alzate Mejía, Nestor
  • Santos Boada, Germán
  • Almeida Amazonas, José Roberto de
This paper is an extended version of our paper published in “Néstor Alzate Mejía, Germán Santos-Boada, José R. de Almeida Amazonas, Decision-Making Under Uncertainty for the Deployment of Future Networks in IoT Scenarios” 11th EAI International Conference on Broadband Communications, Networks, and Systems, 11–12 December 2020., Among the several emerging dimensioning, control and deployment of future communication network paradigms stands out the human-centric characteristic that creates an intricate relationship between telematics and human activities. The hard to model dynamics of user behavior introduces new uncertainties into these systems that give rise to difficult network resource management challenges. According to this context, this work reviews several decision-making computational methods under the influence of uncertainties. This work, by means of a systematic literature review, focuses on sensor-based Internet of Things scenarios such as Smart Spaces and Industry 4.0. According to our conclusions, it is mandatory to establish a means for modeling the human behavior context in order to improve resource assignment and management., This research was funded by Spanish Ministry of Economy and Competitiveness under contract TEC2017-90034-C2-1-R (ALLIANCE project) that received funding from FEDER. Moreover, it was partially supported by the Spanish Thematic Network under contract RED2018-102585-T (Go2Edge) and by the aid granted by the Sinfoni project INV2733 of the Cooperative University of Colombia., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

A hybrid packet/circuit optical transport architecture for DCN

  • Panahi, Nogol
  • Careglio, Davide
  • Solé Pareta, Josep
The aim of this paper is to move away from today's multi-tier, manually operated, and performance limited Data Centre Network (DCN) towards more scalable, flexible, and optimized architecture of tomorrow. We propose a new hybrid optical DCN architecture capable of providing a fully optical interconnection path from server to server within and between racks utilizing fast packet switching at the aggregation layer and high-capacity and low latency circuit switching at the core layer. We have built the DCN control plane upon the Software Defined Networks (SDN) paradigm to leverage the added value of emerging optical technologies for providing flexible and programmable heterogeneous IT and network services. Dynamicity, flexibility, and resiliency are combined in the proposed infrastructure to offer advanced services to the data centre providers in order to fulfil the requirements of the upcoming applications., This work has been partially funded by the Spanish Ministry of Economy and Competitiveness under contract FEDER TEC2017-90034-C2-1-R (ALLIANCE project) and supported but not funded by the Generalitat de Catalunya under contract 2017SGR-1037., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Decentralized trust in the inter-domain routing infrastructure

  • Paillissé Vilanova, Jordi
  • Manrique, Jan
  • Bonet, Guillem
  • Rodríguez Natal, Alberto
  • Maino, Fabio
  • Cabellos Aparicio, Alberto
Inter-domain routing security is of critical importance to the Internet since it prevents unwanted traffic redirections. The current system is based on a Public Key Infrastructure (PKI), a centralized repository of digital certificates. However, the inherent centralization of such design creates tensions between its participants and hinders its deployment. In addition, some technical drawbacks of PKIs delay widespread adoption. In this paper we present IPchain, a blockchain to store the allocations and delegations of IP addresses. IPchain leverages blockchains' properties to decentralize trust among its participants, with the final goal of providing flexible trust models that adapt better to the ever-changing geopolitical landscape. Moreover, we argue that Proof of Stake is a suitable consensus algorithm for IPchain due to the unique incentive structure of this use-case, and that blockchains offer relevant technical advantages when compared to existing systems, such as simplified management. In order to show its feasibility and suitability, we have implemented and evaluated IPchain's performance and scalability storing around 350k IP prefixes in a 2.5 GB chain., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

SD-access: practical experiences in designing and deploying software defined enterprise networks

  • Paillissé Vilanova, Jordi
  • Portoles Comeras, Marc
  • López Brescó, Albert
  • Rodríguez Natal, Alberto
  • Iacobacci, David
  • Leong, Johnson
  • Moreno, Víctor
  • Cabellos Aparicio, Alberto
  • Maino, Fabio
  • Hooda, Sanjay
Enterprise networks, over the years, have become more and more complex trying to keep up with new requirements that challenge traditional solutions. Just to mention one out of many possible examples, technologies such as Virtual LANs (VLANs) struggle to address the scalability and operational requirements introduced by Internet of Things (IoT) use cases. To keep up with these challenges we have identified four main requirements that are common across modern enterprise networks: (i) scalable mobility, (ii) endpoint segmentation, (iii) simplified administration, and (iv) resource optimization. To address these challenges we designed SDA (Software Defined Access), a solution for modern enterprise networks that leverages Software-Defined Networking (SDN) and other state of the art techniques. In this paper we present the design, implementation and evaluation of SDA. Specifically, SDA: (i) leverages a combination of an overlay approach with an event-driven protocol (LISP) to dynamically adapt to traffic and mobility patterns while preserving resources, and (ii) enforces policies to groups of endpoints for scalable segmentation with low operational burden. We present our experience with deploying SDA in two real-life scenarios: an enterprise campus, and a large warehouse with mobile robots. Our evaluation shows that SDA, when compared with traditional enterprise networks, can (i) reduce overall data plane forwarding state up to 70% thanks to a reactive protocol using a centralized routing server, and (ii) reduce by an order of magnitude the handover delays in scenarios of massive mobility with respect to other approaches. Finally, we discuss lessons learned while deploying and operating SDA, and possible optimizations regarding the use of an event-driven protocol and group-based segmentation., This work was partially supported by the Spanish MINECO under contract TEC2017-90034-C2-1-R (ALLIANCE) and 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)

Towards accurate classification of HTTPS traffic in Software-Defined Networks

  • Suárez-Varela Maciá, José Rafael
  • Barlet Ros, Pere
In nowadays Internet, there is a strong trend to encrypt the traffic in order to protect users' privacy. This results in a hard challenge for traffic classification, as the payload in the packets cannot be accessed anymore. In this context, some techniques were proposed for traditional networks in order to classify this traffic. However, we could not find previous works addressing encrypted traffic classification considering the particularities of the Software-Defined Networking (SDN) paradigm. In this paper we present an OpenFlow-based classification system which combines techniques leveraging information in SSL/TLS certificates and DNS traffic to perform accurate flow-level classification for encrypted traffic. We make experiments with real-world traffic to evaluate the overall classification accuracy of our system as well as the accuracy detecting specific popular applications (e.g., Netflix). Furthermore, we assess the processing overhead when deploying our system in SDN environments. As a result, we observe that the support provided by OpenFlow enables to achieve a high accuracy with a more reduced processing overhead than in traditional networks, where typically the whole traffic is mirrored to an external collector that classifies the traffic., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Evaluation of Add/Drop flexibility requirements in future SDM-enabled ROADMs

  • Perelló Muntan, Jordi
  • Gené Bernaus, Joan M.
  • Klinkowski, Miroslaw
  • Comellas Colomé, Jaume
  • Spadaro, Salvatore
In this paper, we evaluate five different architectures for implementing the Add/Drop module of future Spatial Division Multiplexing (SDM)-enabled Re-configurable Optical Add & Drop Multiplexers (ROADMs) with different degrees of flexibility, from a completely static transponder to spatial channel assignment to a fully-flexible transceiver to fibre and spatial channel assignment. The performance of the proposed architectures is compared in a single SDM-enabled ROADM scenario, where local as well as bypass traffic is injected, supported by spectral super-channels at diverse bit-rates and employing different modulation formats. The obtained results illustrate the additional benefits attainable by more flexible Add/Drop architectures versus their increased hardware complexity., The work included in this paper has been supported by the Spanish Government through project ALLIANCE (TEC2017-90034-C2-1-R and TEC2017-90034-C2-2-R) with FEDER contribution. The work of M. Klinkowski was supported by National Science Centre, Poland under Grant 2016/21/B/ST7/02212., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Towards more realistic network models based on Graph Neural Networks

  • Badia Sampera, Arnau
  • Suárez-Varela Maciá, José Rafael
  • Almasan Puscas, Felician Paul
  • Rusek, Krzysztof
  • Barlet Ros, Pere
  • Cabellos Aparicio, Alberto
Recently, a Graph Neural Network (GNN) model called RouteNet was proposed as an efficient method to estimate end-to-end network performance metrics such as delay or jitter, given the topology, routing, and traffic of the network. Despite its success in making accurate estimations and generalizing to unseen topologies, the model makes some simplifying assumptions about the network, and does not consider all the particularities of how real networks operate. In this work we extend the architecture of RouteNet to support different features on forwarding devices, specifically we focus on devices with variable queue sizes, and we experimentally evaluate the accuracy of the extended RouteNet architecture., This work was supported by the Spanish MINECO under contract TEC2017-90034-C2-1-R (ALLIANCE), the Catalan Institution for Research and Advanced Studies (ICREA), FI-AGAUR grant by the Catalan Government and the AGH University of Science and by the Polish Ministry of Science and Higher Education with the subvention funds of the Faculty of Computer Science, Electronics and Telecommunications of AGH University. The research was also supported in part by PL-Grid Infrastructure., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Immersive interconnected virtual and augmented reality: a 5G and IoT perspective

  • Torres Vega, Maria
  • Liaskos, Christos
  • Abadal Cavallé, Sergi
  • Papapetrou, Evangelos
  • Jain, Akshay
  • Mouhouche, Belkacem
  • Kalem, Gokhan
  • Ergüt, Salih
  • Mach, Marian
  • Sabol, Tomas
  • Cabellos Aparicio, Alberto
  • Grimm, Christoph
  • De Turck, Filip
  • Famaey, Jeroen
The final publication is available at Springer via http://dx.doi.org/10.1007/s10922-020-09545-w, Despite remarkable advances, current augmented and virtual reality (AR/VR) applications are a largely individual and local experience. Interconnected AR/VR, where participants can virtually interact across vast distances, remains a distant dream. The great barrier that stands between current technology and such applications is the stringent end-to-end latency requirement, which should not exceed 20 ms in order to avoid motion sickness and other discomforts. Bringing AR/VR to the next level to enable immersive interconnected AR/VR will require significant advances towards 5G ultra-reliable low-latency communication (URLLC) and a Tactile Internet of Things (IoT). In this article, we articulate the technical challenges to enable a future AR/VR end-to-end architecture, that combines 5G URLLC and Tactile IoT technology to support this next generation of interconnected AR/VR applications. Through the use of IoT sensors and actuators, AR/VR applications will be aware of the environmental and user context, supporting human-centric adaptations of the application logic, and lifelike interactions with the virtual environment. We present potential use cases and the required technological building blocks. For each of them, we delve into the current state of the art and challenges that need to be addressed before the dream of remote AR/VR interaction can become reality., Maria Torres Vega is funded by the Research Foundation Flanders (FWO), grant number 12W4819N. This work has been partially supported by the Spanish Ministry of Economy and Competitiveness under contract TEC2017-90034-C2-1-R (ALLIANCE project) that receives funding from FEDER, Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Flexible resource provisioning of polarization independent coherent PONs based on non-orthogonal multiple access and multiCAP modulation

  • Izquierdo Nuñez, David
  • Sarmiento Hernández, Samael
  • Barrio Segura, Miguel
  • Clemente Vicente, Jesús
  • Millán, Pablo
  • Lázaro Villa, José Antonio
  • Garces Gregorio, Juan Ignacio
  • Altabas Navarro, José Antonio
  • Puerta, Rafael
  • Rommel, Simon
  • Vegas Olmos, Juan J.
  • Tafur Monroy, Idelfonso
A combined non-orthogonal multiple access (NOMA) and multiband carrierless amplitude and phase modulation (multiCAP) scheme is proposed for flexible resource provisioning in coherent passive optical networks (PONs). While the proposed combination increases the data-rate and the number of users through the inclusion of NOMA and multiCAP, the coherent reception increases the range and splitting factor of the network through optical power budget enhancement. The proposed system is experimentally evaluated providing a 20 Gb/s aggregated data-rate per wavelength with 10G optoelectronics and applied in two main PON scenarios with four spectral configurations from full-band CAP to four-band multiCAP. The first PON scenario consists of an existing PON that the network operator requires to increase the number of users; NOMA with full-band CAP or multiCAP can be introduced replacing one or more users of the existing PON by one or several new sub-networks and multiplexing the sub-networks by NOMA. In the other PON scenario, several PONs can be nested sharing an initial splitter and achieving a larger number of users by the combination of NOMA and multiCAP. The proposed techniques are fully compatible with other multiplexing techniques such as dense wavelength division multiplexing, which should be considered to achieve higher numbers of users., Gobierno de Aragón (T20_20R); Ministerio de Economía y Competitividad (TEC2017-85752-R, TEC2017-90034-C2-2-R); Electronic Components and Systems for European Leadership (H2020/ECSEL BRAINE - 876967)., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Decentralised Internet infrastructure: Securing inter-domain routing (DEMO)

  • Ferriol Galmés, Miquel
  • Cabellos Aparicio, Alberto
The Border Gateway Protocol (BGP) is the inter-domain routing protocol that glues the Internet. BGP does not incorporate security and instead, it relies on careful configuration and manual filtering to offer some protection. As a consequence, the current inter-domain routing infrastructure is partially vulnerable to prefix and path hijacks as well as in misconfigurations that results in route leaks. There are many instances of these vulnerabilities being exploited by malicious actors on the Internet, resulting in disruption of services. To address this issue the IETF has designed RPKI, a centralised trust architecture that relies on Public Key Infrastructure. RPKI has slow adoption and its centralised nature is problematic: network administrators are required to trust CAs and do not have the ultimate control of their own critical Internet resources (e.g,. IP blocks, AS Numbers). In this context, we have built the Decentralised Internet Infrastructure (DII), a distributed ledger to securely store inter-domain routing information. The main advantages of DII are (i) it offers flexible trust models where the Internet community can define the rules of a consensus algorithm that properly reflects the power balance of its members and, (ii) offers protection against vulnerabilities (path hijack and route leaks) that goes well beyond what RPKI offers. We have deployed the prototype on the wild in a worldwide testbed including 7 ASes, we will use the testbed to demonstrate in a realistic scenario how allocation and delegation of Internet resources in DII work, and how this protects ASes against artificially produced path and prefix hijack as well as a route leak., This work was partially supported by the Spanish MINECO under contract TEC2017-90034-C2-1-R (ALLIANCE) and 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)

Experimenting with real application-specific QoS guarantees in a large-scale RINA demonstrator

  • Perelló Muntan, Jordi
  • López Brescó, Albert
  • Careglio, Davide
© 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., This paper reports the definition, setup and obtained results of the Fed4FIRE + medium experiment ERASER, aimed to evaluate the actual Quality of Service (QoS) guarantees that the clean-slate Recursive InterNetwork Architecture (RINA) can deliver to heterogeneous applications at largescale. To this goal, a 37Node 5G metro/regional RINA network scenario, spanning from the enduser to the server where applications run in a datacenter has been configured in the Virtual Wall experimentation facility. This scenario has initially been loaded with synthetic application traffic flows, with diverse QoS requirements, thus reproducing different network load conditions. Next,their experienced QoS metrics endtoend have been measured with two different QTAMux (i.e., the most accepted candidate scheduling policy for providing RINA with its QoS support) deployment scenarios. Moreover, on this RINA network scenario loaded with synthetic application traffic flows, a real HD (1080p) video streaming demonstration has also been conducted, setting up video streaming sessions to endusers at different network locations, illustrating the perceived Quality of Experience (QoE). Obtained results in ERASER disclose that, by appropriately deploying and configuring QTAMux, RINA can yield effective QoS support, which has provided perfect QoE in almost all locations in our demo when assigning video traffic flows the highest (i.e., Gold) QoS Cube., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Joint latency and reliability-aware controller placement

  • Rasol, Kurdman Abdulrahman Rasol
  • Domingo Pascual, Jordi
In network architectures based on Software Defined Networking (SDN) the control plane (control logic) is separated from the network data plane (forwarding plane) while traditional network routers combine both. Software Defined networks facilitates a centralized networking system where a logical controller manages the global view of the network. In this paper, we first propose a new metric on the controller placement problem (CPP) that simultaneously considers the communication latency and communication reliability both between switches and controllers and between controllers. Reliability is considered for single-link failure. We model the problem of determining the optimal controller placement to provide low latencies in the control plane traffic. The objective of this study is to minimize the average accumulated latency by jointly taking into account the latency between controller to switches and inter-controller while optimizing their placement for achieving an optimal balance simultaneously. The optimization problem is formulated as a mixed-integer linear programming (MILP) model under the constraints of latency and reliability. We evaluated the performance of our proposed metric by using the Internet2 OS3E network topology. Different from previous work, we focus on the control traffic exchanged among controllers to synchronize their shared data structure. Results demonstrate that the proposed method is promising., This research was supported by the Spanish Ministry of Economy and Competitiveness under contract TEC2017-90034-C2-1-R (ALLIANCE)., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Routing in optical transport networks with deep reinforcement learning

  • Suárez Varela, José
  • Mestres Sugranyes, Albert
  • Yu, Junlin
  • Kuang, Li
  • Feng, Haoyu
  • Cabellos Aparicio, Alberto
  • Barlet Ros, Pere
Deep reinforcement learning (DRL) has recently revolutionized the resolution of decision-making and automated control problems. In the context of networking, there is a growing trend in the research community to apply DRL algorithms to optimization problems such as routing. However, existing proposals fail to achieve good results, often under-performing traditional routing techniques. We argue that the reason behind this poor performance is that they use straightforward representations of networks. In this paper, we propose a DRL-based solution for routing in optical transport networks (OTNs). Contrary to previous works, we propose a more elaborate representation of the network state that reduces the level of knowledge abstraction required for DRL agents and easily captures the singularities of network topologies. Our evaluation results show that using our novel representation, DRL agents achieve better performance and learn how to route traffic in OTNs significantly faster compared to state-of-the-art representations. Additionally, we reverse engineered the routing strategy learned by our DRL agent, and as a result, we found a routing algorithm that outperforms well-known traditional routing heuristics., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Results and achievements of the ALLIANCE Project: New network solutions for 5G and beyond

  • Careglio, Davide
  • Spadaro, Salvatore
  • Cabellos Aparicio, Alberto
  • Lázaro Villa, José Antonio
  • Barlet Ros, Pere
  • Gené Bernaus, Joan M.
  • Perelló Muntan, Jordi
  • Agraz Bujan, Fernando
  • Suárez-Varela Maciá, José Rafael
  • Pagès Raventós, Albert
  • Paillissé Vilanova, Jordi
  • Almasan Puscas, Felician Paul
  • Domingo Pascual, Jordi
  • Solé Pareta, Josep
Leaving the current 4th generation of mobile communications behind, 5G will represent a disruptive paradigm shift integrating 5G Radio Access Networks (RANs), ultra-high-capacity access/metro/core optical networks, and intra-datacentre (DC) network and computational resources into a single converged 5G network infrastructure. The present paper overviews the main achievements obtained in the ALLIANCE project. This project ambitiously aims at architecting a converged 5G-enabled network infrastructure satisfying those needs to effectively realise the envisioned upcoming Digital Society. In particular, we present two networking solutions for 5G and beyond 5G (B5G), such as Software Defined Networking/Network Function Virtualisation (SDN/NFV) on top of an ultra-high-capacity spatially and spectrally flexible all-optical network infrastructure, and the clean-slate Recursive Inter-Network Architecture (RINA) over packet networks, including access, metro, core and DC segments. The common umbrella of all these solutions is the Knowledge-Defined Networking (KDN)-based orchestration layer which, by implementing Artificial Intelligence (AI) techniques, enables an optimal end-to-end service provisioning. Finally, the cross-layer manager of the ALLIANCE architecture includes two novel elements, namely the monitoring element providing network and user data in real time to the KDN, and the blockchain-based trust element in charge of exchanging reliable and confident information with external domains., This work has been partially funded by the Spanish Ministry of Economy and Competitiveness under contract FEDER TEC2017-90034-C2 (ALLIANCE project) and by the Generalitat de Catalunya under contract 2017SGR-1037 and 2017SGR-605., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Using spatial division multiplexing to avoid fragmentation in flex-grid optical networks

  • Comellas Colomé, Jaume
  • Perelló Muntan, Jordi
  • Solé Pareta, Josep
  • Junyent Giralt, Gabriel
Spectrum fragmentation is one of the main drawbacks of Flex-Grid Optical Networks, and it becomes especially harmful under dynamic traffic conditions. When evolving from single fibre networks to spatial division multiplexed (SDM) ones, where many parallel cores/fibres are deployed, spectrum fragmentation still limits their eventual performance. In this work, the total available spectrum is partitioned according to the SDM dimension, and each one of the resulting partitions is assigned to each type of connections. This strategy is proposed as a simple method to eliminate spectrum fragmentation. By allocating all same sized connections onto the same spectral/spatial partition, the randomness in the size of the spectral voids left by torn down connections is eliminated. The proposal is numerically evaluated using different network topologies. The obtained results show that spatial partitioning is an interesting solution for spectrum fragmentation avoidance in these scenarios., This work has been partially funded by Spanish projects TWINS (TEC2017-90097-R) and ALLIANCE (TEC2017-90034-C2-1-R)., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

RouteNet: leveraging graph neural networks for network modeling and optimization in SDN

  • Rusek, Krzysztof
  • Suárez-Varela Maciá, José Rafael
  • Almasan Puscas, Felician Paul
  • Barlet Ros, Pere
  • Cabellos Aparicio, Alberto
© 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., Network modeling is a key enabler to achieve efficient network operation in future self-driving Software-Defined Networks. However, we still lack functional network models able to produce accurate predictions of Key Performance Indicators (KPI) such as delay, jitter or loss at limited cost. In this paper we propose RouteNet, a novel network model based on Graph Neural Network (GNN) that is able to understand the complex relationship between topology, routing, and input traffic to produce accurate estimates of the per-source/destination per-packet delay distribution and loss. RouteNet leverages the ability of GNNs to learn and model graph-structured information and as a result, our model is able to generalize over arbitrary topologies, routing schemes and traffic intensity. In our evaluation, we show that RouteNet is able to predict accurately the delay distribution (mean delay and jitter) and loss even in topologies, routing and traffic unseen in the training (worst case MRE = 15.4%). Also, we present several use cases where we leverage the KPI predictions of our GNN model to achieve efficient routing optimization and network planning., This work was supported in part by the Polish Ministryof Science and Higher Education with the subvention funds of the Facultyof Computer Science, Electronics and Telecommunications, AGH University,in part by the Spanish MINECO under Contract TEC2017-90034-C2-1-R(ALLIANCE), in part by the Catalan Institution for Research and AdvancedStudies (ICREA) and the FI-AGAUR Grant by the Catalan Government, andin part by PL-Grid Infrastructure., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

On the benefits of probabilistic constellation shaping in Flex-Grid/MCF dynamic optical backbone networks

  • Perelló Muntan, Jordi
  • Gené Bernaus, Joan M.
  • Spadaro, Salvatore
In this paper, we propose a novel methodology for estimating the worst-case spectral efficiency (b/s/Hz) of end-to-end paths across a Multi-Core Fiber (MCF)-enabled Flex-Grid optical network, accounting for physical link and intra-node noise, together with the inter-core crosstalk impairment appearing during the transmission of optical signals over MCFs. In particular, we employ the proposed methodology to quantify the benefits of Probabilistic Constellation Shaping (PCS) versus traditional polarization-multiplexed modulation formats in two reference backbone networks. From the obtained results, PCS yields admissible offered load gains close to 13% under 1% bandwidth blocking probability., This work has been supported by the Spanish Government through project ALLIANCE (TEC2017-90034-C2-1-R and TEC2017-90034-C2-2-R) with FEDER contribution. Moreover, it has been supported by the Spanish Thematic Network under contract RED2018-102585-T (Go2Edge)., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

A control plane for WireGuard

  • Paillissé Vilanova, Jordi
  • Barcia González, Alejandro
  • López Brescó, Albert
  • Rodríguez Natal, Alberto
  • Maino, Fabio
  • Cabellos Aparicio, Alberto
WireGuard is a VPN protocol that has gained significant interest recently. Its main advantages are: (i) simple configuration (via pre-shared SSH-like public keys), (ii) mobility support, (iii) reduced codebase to ease auditing, and (iv) Linux kernel implementation that yields high performance. However, WireGuard (intentionally) lacks a control plane. This means that each peer in a WireGuard network has to be manually configured with the other peers’ public key and IP addresses, or by other means. In this paper we present an architecture based on a centralized server to automatically distribute this information. In a nutshell, first we manually establish a WireGuard tunnel to the centralized server, and ask all the peers to store their public keys and IP addresses in it. Then, WireGuard peers use this secure channel to retrieve on-demand the information for the peers they want to communicate to. Our design strives to: (i) offer a key distribution scheme simpler than PKI-based ones, (ii) limit the number of public keys sent to the peers, and (iii) reduce tunnel establishment latency by means of an UDP-based protocol. We argue that such automation can help the deployment in enterprise or ISP scenarios. We also describe in detail our implementation and analyze several performance metrics. Finally, we discuss possible improvements regarding several shortcomings we found during implementation., This work was partially supported by the Spanish MINECO under contract TEC2017-90034-C2-1-R (ALLIANCE) and 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)

Wide area network autoscaling for cloud applications

  • Serracanta Pujol, Berta
  • Paillissé Vilanova, Jordi
  • Cabellos Aparicio, Alberto
  • Claiborne, Anna
  • Rodríguez Natal, Alberto
  • Ward, Dave
  • Maino, Fabio
Modern cloud orchestrators like Kubernetes provide a versatile and robust way to host applications at scale. One of their key features is autoscaling, which automatically adjusts cloud resources (compute, memory, storage) in order to adapt to the demands of applications. However, the scope of cloud autoscaling is limited to the datacenter hosting the cloud and it doesn't apply uniformly to the allocation of network resources. In I/O-constrained or data-in-motion use cases this can lead to severe performance degradation for the application. For example, when the load on a cloud service increases and the Wide Area Network (WAN) connecting the datacenter to the Internet becomes saturated, the application flows experience an increase in delay and loss. In many cases this is dealt with overprovisioning network capacity, which introduces additional costs and inefficiencies. On the other hand, thanks to the concept of "Network as Code", the WAN exposes a set of APIs that can be used to dynamically allocate and de-allocate capacity on-demand. In this paper we propose extending the concept of cloud autoscaling into the network to address this limitation. This way, applications running in the cloud can communicate their networking requirements, like bandwidth or traffic profile, to a Software-Defined Networking (SDN) controller or Network as a Service (NaaS) platform. Moreover, we aim to define the concepts of vertical and horizontal autoscaling applied to networking. We present a prototype that automatically allocates bandwidth to the underlay network, according to the requirements of the applications hosted in Kubernetes. Finally, we discuss open research challenges., This work was supported by the Spanish MINECO under contract TEC2017-90034-C2-1-R (ALLIANCE), the Catalan Institution for Research and Advanced Studies (ICREA)., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

The graph neural networking challenge: a worldwide competition for education in AI/ML for networks

  • Suárez-Varela Maciá, José Rafael
  • Ferriol Galmés, Miquel
  • López Brescó, Albert
  • Almasan Puscas, Felician Paul
  • Bernárdez Gil, Guillermo
  • Pujol Perich, David
  • Rusek, Krzysztof
  • Bonniot, Loïck
  • Neumann, Christoph
  • Schnitzler, François
  • Taïani, François
  • Happ, Martin
  • Barlet Ros, Pere
  • Cabellos Aparicio, Alberto
During the last decade, Machine Learning (ML) has increasingly become a hot topic in the field of Computer Networks and is expected to be gradually adopted for a plethora of control, monitoring and management tasks in real-world deployments. This poses the need to count on new generations of students, researchers and practitioners with a solid background in ML applied to networks. During 2020, the International Telecommunication Union (ITU) has organized the "ITU AI/ML in 5G challenge", an open global competition that has introduced to a broad audience some of the current main challenges in ML for networks. This large-scale initiative has gathered 23 different challenges proposed by network operators, equipment manufacturers and academia, and has attracted a total of 1300+ participants from 60+ countries. This paper narrates our experience organizing one of the proposed challenges: the "Graph Neural Networking Challenge 2020". We describe the problem presented to participants, the tools and resources provided, some organization aspects and participation statistics, an outline of the top-3 awarded solutions, and a summary with some lessons learned during all this journey. As a result, this challenge leaves a curated set of educational resources openly available to anyone interested in the topic., This work has received funding from the European Union’s H2020 research and innovation programme within the framework of the NGI-POINTER Project funded under grant agreement No. 871528. This paper reflects only the authors’ view; the European Commission is not responsible for any use that may be made of the information it contains. This work was also supported by the Spanish MINECO under contract TEC2017-90034-C2-1-R (ALLIANCE), the Catalan Institution for Research and Advanced Studies (ICREA), and by FI-AGAUR grant by the Catalan Government. Salzburg Research is grateful for the support by the WISS 2025 (Science and Innovation Strategy Salzburg 2025) project ”IDALab Salzburg” (20204-WISS/225/197-2019 and 20102-F1901166-KZP) and the 5G-AI-MLab by the Federal Ministry for Climate Action, Environment, Energy, Mobility, Innovation and Technology (BMK) and the Austrian state Salzburg., Peer Reviewed, Article escrit per 24 autors/autores: José Suárez-Varela (1), Miquel Ferriol-Galmés (1), Albert López (1), Paul Almasan (1), Guillermo Bernárdez (1), David Pujol-Perich (1), Krzysztof Rusek (1,2), Loïck Bonniot (3, 4), Christoph Neumann (3), François Schnitzler (3), François Taïani (4), Martin Happ (5, 6), Christian Maier (5), Jia Lei Du (5), Matthias Herlich (5), Peter Dorfinger (5), Nick Vincent Hainke (7), Stefan Venz (7), Johannes Wegener (7), Henrike Wissing (7), Bo Wu (8), Shihan Xiao (8), Pere Barlet-Ros (1), Albert Cabellos-Aparicio (1). 1- Barcelona Neural Networking center, Universitat Politècnica de Catalunya, Spain. 2- AGH University of Science and Technology, Department of Telecommunications, Poland. 3- InterDigital, France. 4- Univ. Rennes, Inria, CNRS, IRISA, France. 5- Salzburg Research Forschungsgesellschaft mbH, Austria. 6- IDA Lab, University of Salzburg, Austria. 7- Fraunhofer HHI, Germany. 8- Network Technology Lab., Huawei Technologies Co., Ltd., China., Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Towards real-time routing optimization with deep reinforcement learning: open challenges

  • Almasan Puscas, Felician Paul
  • Suárez-Varela Maciá, José Rafael
  • Wu, Bo
  • Xiao, Shihan
  • Barlet Ros, Pere
  • Cabellos Aparicio, Alberto
The digital transformation is pushing the existing network technologies towards new horizons, enabling new applications (e.g., vehicular networks). As a result, the networking community has seen a noticeable increase in the requirements of emerging network applications. One main open challenge is the need to accommodate control systems to highly dynamic network scenarios. Nowadays, existing network optimization technologies do not meet the needed requirements to effectively operate in real time. Some of them are based on hand-crafted heuristics with limited performance and adaptability, while some technologies use optimizers which are often too time-consuming. Recent advances in Deep Reinforcement Learning (DRL) have shown a dramatic improvement in decision-making and automated control problems. Consequently, DRL represents a promising technique to efficiently solve a variety of relevant network optimization problems, such as online routing. In this paper, we explore the use of state-of-the-art DRL technologies for real-time routing optimization and outline some relevant open challenges to achieve production-ready DRL-based solutions., This work has received funding from the European Union’s Horizon 2020 research and innovation programme within the framework of the NGI-POINTER Project funded under grant agreement No 871528. This paper reflects only the author’s view; the European Commission is not responsible for any use that may be made of the information it contains. This work was also supported by the Spanish MINECO under contract TEC2017-90034-C2-1-R (ALLIANCE), the Catalan Institution for Research and Advanced Studies (ICREA) and the Secretariat for Universities and Research of the Ministry of Business and Knowledge of the Government of Catalonia and the European Social Fund., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

NetXplain: Real-time explainability of graph neural networks applied to networking

  • Pujol Perich, David
  • Suárez-Varela Maciá, José Rafael
  • Xiao, Shihan
  • Wu, Bo
  • Cabellos Aparicio, Alberto
  • Barlet Ros, Pere
Recent advancements in Deep Learning (DL) have revolutionized the way we can efficiently tackle complex optimization problems. However, existing DL-based solutions are often considered as black boxes with high inner complexity. As a result, there is still certain skepticism among the networking industry about their practical viability to operate data networks. In this context, explainability techniques have recently emerged to unveil why DL models make each decision. This paper focuses on the explainability of Graph Neural Networks (GNNs) applied to networking. GNNs are a novel DL family with unique properties to generalize over graphs. As a result, they have shown unprecedented performance to solve complex network optimization problems. This paper presents NetXplain, a novel real-time explainability solution that uses a GNN to interpret the output produced by another GNN. In the evaluation, we apply the proposed explainability method to RouteNet, a GNN model that predicts end-to-end QoS metrics in networks. We show that NetXplain operates more than 3 orders of magnitude faster than state-of-the-art explainability solutions when applied to networks up to 24 nodes, which makes it compatible with real-time applications; while demonstrating strong capabilities to generalize to network scenarios not seen during training., This work has been supported by the Spanish MINECO under contract TEC2017‑90034‑C2‑1‑R (ALLIANCE) and 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)

Challenging the generalization capabilities of Graph Neural Networks for network modeling

  • Suárez-Varela Maciá, José Rafael
  • Carol Bosch, Sergi
  • Rusek, Krzysztof
  • Almasan Puscas, Felician Paul
  • Arias Vicente, Marta
  • Barlet Ros, Pere
  • Cabellos Aparicio, Alberto
Today, network operators still lack functional network models able to make accurate predictions of end-to-end Key Performance Indicators (e.g., delay or jitter) at limited cost. Recently a novel Graph Neural Network (GNN) model called RouteNet was proposed as a cost-effective alternative to estimate the per-source/destination pair mean delay and jitter in networks. Thanks to its GNN architecture that operates over graph-structured data, RouteNet revealed an unprecedented ability to learn and model the complex relationships among topology, routing and input traffic in networks. As a result, it was able to make performance predictions with similar accuracy than resource-hungry packet-level simulators even in network scenarios unseen during training. In this demo, we will challenge the generalization capabilities of RouteNet with more complex scenarios, including larger topologies., This work was supported by the Spanish MINECO under contract TEC2017-90034-C2-1-R (ALLIANCE), the Catalan Institution for Research and Advanced Studies (ICREA) and the AGH University of Science and Technology grant, under contract no. 15.11.230.400. The research was also supported in part by PL-Grid Infrastructure., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Route, modulation format, MIMO and spectrum assignment in Flex-Grid/MCF transparent optical core networks

  • Rumipamba Zambrano, Rubén Darío
  • Perelló Muntan, Jordi
  • Spadaro, Salvatore
In this paper, we target an optimal multiple-input multiple-output digital signal processing (MIMO-DSP) assignment to super-channels affected by intercore crosstalk (ICXT) in multicore fiber (MCF) enabled transparent optical core networks. MIMO-DSP undoes ICXT effects, but can be costly with high core density MCFs. Hence, its implementation in the network must be carefully decided. We address our objective as a joint route, modulation format, MIMO and spectrum assignment (RMMSA) problem, for which integer linear programming formulations are provided to optimally solve it in small network scenarios. Moreover, several heuristic approaches are also proposed to solve large-scale problem instances with good accuracy. Their goal is to minimize both network spectral requirements and the amount of MIMO equalized super-channels, taking a crosstalk-free space division multiplexing (SDM) solution as a reference, for example, based on parallel single mode fibers [i.e., a multifiber (MF) scenario]. For our evaluation, we consider several state-of-the-art MCF prototypes and different network topologies. The obtained results, with the considered MCFs, disclose that in national backbone networks, the desirable percentage of super-channels with MIMO equalization to match the performance of an equivalent crosstalk-free SDM solution ranges from 0% to 36, while in continental-wide networks this range raises from 0% to 56%. In addition, in the case of a nonideal MIMO (with a 3 dB/km of crosstalk compensation), such percentages range from 0% to 28% and from 0% to 45% in national and continental-wide backbone networks, respectively, experimenting a performance gap up to 12% with respect to the MF reference scenario., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Distributed access control with blockchain

  • Paillissé Vilanova, Jordi
  • Subirà Nieto, Jordi
  • López Brescó, Albert
  • Rodríguez Natal, Alberto
  • Ermagan, Vina
  • Maino, Fabio
  • Cabellos Aparicio, Alberto
The specification and enforcement of network-wide policies in a single administrative domain is common in today's networks and considered as already resolved. However, this is not the case for multi-administrative domains, e.g. among different enterprises. In such situation, new problems arise that challenge classical solutions such as PKIs, which suffer from scalability and granularity concerns. In this paper, we present an extension to Group-Based Policy -a widely used network policy languagefor the aforementioned scenario. To do so, we take advantage of a permissioned blockchain implementation (Hyperledger Fabric) to distribute access control policies in a secure and auditable manner, preserving at the same time the independence of each organization. Network administrators specify polices that are rendered into blockchain transactions. A LISP control plane (RFC 6830) allows routers performing the access control to query the blockchain for authorizations. We have implemented an end-to-end experimental prototype and evaluated it in terms of scalability and network latency., This work has been supported by the Spanish MINECO under contract TEC2017-90034-C2-1-R (ALLIANCE) and 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)

Unveiling the potential of Graph Neural Networks for network modeling and optimization in SDN

  • Rusek, Krzysztof
  • Suárez-Varela Maciá, José Rafael
  • Mestres Sugrañes, Albert
  • Barlet Ros, Pere
  • Cabellos Aparicio, Alberto
Network modeling is a critical component for building self-driving Software-Defined Networks, particularly to find optimal routing schemes that meet the goals set by administrators. However, existing modeling techniques do not meet the requirements to provide accurate estimations of relevant performance metrics such as delay and jitter. In this paper we propose a novel Graph Neural Network (GNN) model able to understand the complex relationship between topology, routing and input traffic to produce accurate estimates of the per-source/destination pair mean delay and jitter. GNN are tailored to learn and model information structured as graphs and as a result, our model is able to generalize over arbitrary topologies, routing schemes and variable traffic intensity. In the paper we show that our model provides accurate estimates of delay and jitter (worst case R2 = 0.86) when testing against topologies, routing and traffic not seen during training. In addition, we present the potential of the model for network operation by presenting several use-cases that show its effective use in per-source/destination pair delay/jitter routing optimization and its generalization capabilities by reasoning in topologies and routing schemes not seen during training., This work was supported by AGH University of Science and Technology grant, under contract no. 15.11.230.400, the Spanish MINECO under contract TEC2017-90034-C2-1-R (ALLIANCE) and the Catalan Institution for Research and Advanced Studies (ICREA). The research was also supported in part by PL-Grid Infrastructure., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

IGNNITION: Bridging the gap between graph neural networks and networking systems

  • Pujol Perich, David
  • Suárez-Varela Maciá, José Rafael
  • Ferriol Galmés, Miquel
  • Xiao, Shihan
  • Wu, Bo
  • Cabellos Aparicio, Alberto
  • Barlet Ros, Pere
Recent years have seen the vast potential of graph neural networks (GNN) in many fields where data is structured as graphs (e.g., chemistry, recommender systems). In particular, GNNs are becoming increasingly popular in the field of networking, as graphs are intrinsically present at many levels (e.g., topology, routing). The main novelty of GNNs is their ability to generalize to other networks unseen during training, which is an essential feature for developing practical machine learning (ML) solutions for networking. However, implementing a functional GNN prototype is currently a cumbersome task that requires strong skills in neural network programming. This poses an important barrier to network engineers that often do not have the necessary ML expertise. In this article, we present IGNNITION, a novel open source framework that enables fast prototyping of GNNs for networking systems. IGNNITION is based on an intuitive high-level abstraction that hides the complexity behind GNNs, while still offering great flexibility to build custom GNN architectures. To showcase the versatility and performance of this framework, we implement two state-of-the-art GNN models applied to different networking use cases. Our results show that the GNN models produced by IGNNITION are equivalent in terms of accuracy and performance to their native implementations in TensorFlow., This open source project has received funding from the European Union’s Horizon 2020 research and innovation programme within the framework of the NGI-POINTER Project funded under grant agreement No. 871528. This article reflects only the author’s view; the European Commission is not responsible for any use that may be made of the information it contains. The work was also supported by the Spanish MINECO under contract TEC2017-90034-C2-1-R (ALLIANCE) and the Catalan Institution for Research and Advanced Studies (ICREA), Peer Reviewed, Postprint (author's final draft)