1,721,021 research outputs found
Towards ML-based Management of Software-Defined Networks
No full textAvec la croissance exponentielle des performances technologiques, le monde moderne est devenu hautement connecté, numérisé et diversifié. Dans ce monde hyperconnecté, les réseaux informatiques ou Internet font partie de notre vie quotidienne et jouent de nombreux rôles importants. Cependant, la forte croissance des services et des applications Internet, ainsi que l'augmentation massive du trafic, complexifient les réseaux qui atteignent un point où les fonctions de gestion traditionnelles, principalement régies par des opérations humaines, ne parviennent pas à maintenir le réseau opérationnel. Dans ce contexte, le Software Defined-Networking (SDN) émerge comme une nouvelle architecture pour la gestion des réseaux. Il rend les réseaux programmables en apportant de la flexibilité dans leur contrôle et leur gestion. Même si la gestion des réseaux est en partie simplifiée, elle reste délicate à cause de la croissance continue de la complexité des réseaux. Les tâches de gestion restent alors complexes. Face à ce constat, le concept de self-driving networking a vu jour. Il consiste à tirer parti des récentes avancées technologiques et l'innovation scientifique dans le domaine de l'intelligence artificielle (IA) et du machine learning (ML) en complément au SDN. Par rapport aux approches de gestion traditionnelles utilisant uniquement des modèles mathématiques analytiques et l'optimisation, ce nouveau paradigme est une approche axée sur les données. Les opérations de gestion s'appuieront sur la capacité de l'intelligence artificielle à exploiter les relations complexes et cachées dans les données pour créer des connaissances. Cette association SDN-AI/ML, avec la promesse de simplifier la gestion du réseau, nécessite de relever de nombreux défis. Le self-driving networking ou l'automatisation complète du réseau est le "Saint Graal" de cette association. Dans cette thèse, deux des défis concernés retiennent notre attention. Dans un premier temps, la collecte efficace de données avec SDN, en particulier la télémétrie en temps réel. Pour ce défi, nous proposons COCO pour COnfidence-based COllection, une solution de collecte de données en temps quasi-réel à faible coût (overhead) pour les réseaux SDN. Les données d'intérêt sont collectées efficacement du plan de données au plan de contrôle, où elles sont utilisées par les applications de gestion traditionnelles ou par des algorithmes de machine learning. Dans un second temps, nous explorons les possibilités de l'utilisation du machine learning pour traiter des tâches de gestion complexes. Nous considérons l'optimisation de la performance des trafics dans les data centers. Nous proposons un modèle de performance du trafic incast en utilisant le machine learning, là où les modèles analytiques peinent à fournir des modèles de performance facilement généralisables et sans des connaissances "domain-specific". Avec ce modèle de performance ML, des fonctions de management dont la gestion intelligente des switchs ou d'autres algorithmes d'optimisation de la qualité de service peuvent optimiser dynamiquement les performances des trafics. Ces opérations de gestion basées sur le ML sont construites sur une architecture SDN, en tirant parti de sa vision globale centralisée, de ses capacités de monitoring et de sa flexibilité. L'efficacité de notre proposition de collecte de données et l'optimisation des performances basée sur le machine learning donnent des résultats prometteurs. Nous pensons que des systèmes de monitoring SDN efficaces couplés avec les opportunités offertes par l'IA/ML peuvent considérablement améliorer la gestion du réseau et faire un grand pas vers le concept du self-driving network et donc des réseaux autonomes.With the exponential growth in technology performance, the modern world has become highly connected, digitized, and diverse. Within this hyper-connected world, Communication networks or the Internet are part of our daily life and play many important roles. However, the ever-growing internet services, application, and massive traffic growth complexify networks that reach a point where traditional management functions mainly govern by human operations fail to keep the network operational. In this context, Software-Defined Networking (SDN) emerge as a new architecture for network management. It makes networks programmable by bringing flexibility in their control and management. Even if network management is eased, it is still tricky to handle due to the continuous growth of network complexity. Management tasks remain then complex. Faced with this, the concept of self-driving networking arose. It consists of leveraging recent technological advancements and scientific innovation in Artificial Intelligence (AI)/Machine Learning (ML) with SDN. Compared to traditional management approaches using only analytic mathematical models and optimization, this new paradigm is a data-driven approach. The management operations will leverage the ML ability to exploit hidden pattern in data to create knowledge. This association SDN-AI/ML, with the promise to simplify network management, needs many challenges to be addresses. Self-driving networking or full network automation is the "Holy Grail" of this association. In this thesis, two of the concerned challenges retain our attention. Firstly, efficient data collection with SDN, especially real-time telemetry. For this challenge, we propose COCO for COnfidence-based COllection, a low overhead near-real-time data collection in SDN. Data of interest is collected efficiently from the data plane to the control plane, where they are used whether by traditional management applications or machine-learning-based algorithms. Secondly, we tackle the effectiveness of the use of machine learning to handle complex management tasks. We consider application performance optimization in data centers. We propose a machine-learning-based incast performance inference, where analytical models struggle to provide general and expert-knowledge-free performance models. With this ML-performance model, smart buffering schemes or other QoS optimization algorithms could dynamically optimize traffic performance. These ML-based management schemes are built upon SDN, leveraging its centralized global view, telemetry capabilities, and management flexibility. The effectiveness of our efficient data collection framework and the machine-learning-based performance optimization show promising results. We expect that improved SDN monitoring with AI/ML analytics capabilities can considerably augment network management and make a big step in the self-driving network journey
Going Beyond Counting First Authors in Author Co-citation Analysis
Full text linkThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
Variations on the Author
Full text link“Variations on the Author” discusses two of Eduardo Coutinho’s recent films (Um Dia na Vida, from 2010, and Últimas Conversas, posthumously released in 2015) and their contribution to the general question of documentary authorship. The director’s filmography is characterized by a consistent yet self-effacing form of authorial self-inscription: Coutinho often features as an interviewer that rather than express opinions propels discourses; an interviewer that is good at listening. This mode of self-inscription characterizes him as an author who is not expressive but who is nonetheless markedly present on the screen. In Um Dia na Vida, however, Coutinho is completely absent form the image, while Últimas Conversas, on the contrary, includes a confessional prologue that moves the director from the margins to the center of his films. This article examines the ways in which these works stand out in the filmography of a director who offers new insights into the notion of cinematic authorship
Appropriate Similarity Measures for Author Cocitation Analysis
Full text linkWe provide a number of new insights into the methodological discussion about author cocitation analysis. We first argue that the use of the Pearson correlation for measuring the similarity between authors’ cocitation profiles is not very satisfactory. We then discuss what kind of similarity measures may be used as an alternative to the Pearson correlation. We consider three similarity measures in particular. One is the well-known cosine. The other two similarity measures have not been used before in the bibliometric literature. Finally, we show by means of an example that our findings have a high practical relevance.information science;Pearson correlation;cosine;similarity measure;author cocitation analysis
Détection non-supervisée d'anomalies du trafic réseau
No full textLa détection d'anomalies est une tâche critique de l'administration des réseaux. L'apparition continue de nouvelles anomalies et la nature changeante du trafic réseau compliquent de fait la détection d'anomalies. Les méthodes existantes de détection d'anomalies s'appuient sur une connaissance préalable du trafic : soit via des signatures créées à partir d'anomalies connues, soit via un profil de normalité. Ces deux approches sont limitées : la première ne peut détecter les nouvelles anomalies et la seconde requiert une constante mise à jour de son profil de normalité. Ces deux aspects limitent de façon importante l'efficacité des méthodes de détection existantes.\ud
\ud
Nous présentons une approche non-supervisée qui permet de détecter et caractériser les anomalies réseaux de façon autonome. Notre approche utilise des techniques de partitionnement afin d'identifier les flux anormaux. Nous proposons également plusieurs techniques qui permettent de traiter les anomalies extraites pour faciliter la tâche des opérateurs. Nous évaluons les performances de notre système sur des traces de trafic réel issues de la base de trace MAWI. Les résultats obtenus mettent en évidence la possibilité de mettre en place des systèmes de détection d'anomalies autonomes et fonctionnant sans connaissance préalable.-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------Anomaly detection has become a vital component of any network in today’s Internet. Ranging from non-malicious unexpected events such as flash-crowds and failures, to network attacks such as denials-of-service and network scans, network traffic anomalies can have serious detrimental effects on the performance and integrity of the network. The continuous arising of new anomalies and attacks create a continuous challenge to cope with events that put the network integrity at risk. Moreover, the inner polymorphic nature of traffic caused, among other things, by a highly changing protocol landscape, complicates anomaly detection system's task. In fact, most network anomaly detection systems proposed so far employ knowledge-dependent techniques, using either misuse detection signature-based detection methods or anomaly detection relying on supervised-learning techniques. However, both approaches present major limitations: the former fails to detect and characterize unknown anomalies (letting the network unprotected for long periods) and the latter requires training over labeled normal traffic, which is a difficult and expensive stage that need to be updated on a regular basis to follow network traffic evolution. Such limitations impose a serious bottleneck to the previously presented problem.\ud
We introduce an unsupervised approach to detect and characterize network anomalies, without relying on signatures, statistical training, or labeled traffic, which represents a significant step towards the autonomy of networks. Unsupervised detection is accomplished by means of robust data-clustering techniques, combining Sub-Space clustering with Evidence Accumulation or Inter-Clustering Results Association, to blindly identify anomalies in traffic flows. Correlating the results of several unsupervised detections is also performed to improve detection robustness. The correlation results are further used along other anomaly characteristics to build an anomaly hierarchy in terms of dangerousness. Characterization is then achieved by building efficient filtering rules to describe a detected anomaly. The detection and characterization performances and sensitivities to parameters are evaluated over a substantial subset of the MAWI repository which contains real network traffic traces.\ud
Our work shows that unsupervised learning techniques allow anomaly detection systems to isolate anomalous traffic without any previous knowledge. We think that this contribution constitutes a great step towards autonomous network anomaly detection.\ud
This PhD thesis has been funded through the ECODE project by the European Commission under the Framework Programme 7. The goal of this project is to develop, implement, and validate experimentally a cognitive routing system that meet the challenges experienced by the Internet in terms of manageability and security, availability and accountability, as well as routing system scalability and quality. The concerned use case inside the ECODE project is network anomaly detection.\ud
\ud
\u
Control design for energy aware communication networks
No full textLes outils informatiques (comme les routeurs et calculateurs entre autres) sont des consommateurs accrus d'énergie. Cette problématique a été déjà prise en compte dans les réseaux mobiles. La question de l'énergie commence juste à être prise en compte pour les systèmes "fixes" à grande échelle qui atteignent de nos jours des tailles impressionnantes. L'objectif de cette thèse est de traiter le problème de la consommation de l'énergie dans les réseaux de communication filaires: fournir un certain niveau de qualité de service (QdS) par rapport à la perte des paquets, la vitesse de réponse et la robustesse par rapport aux différentes périodes d'échantillonnages tout en contrôlant la puissance consommée du système. Le but est de concevoir une méthode à partir de la théorie de la commande, qui consiste à garantir un certain nombre de paramètres de QdS. Cette technique est appliquée au niveau local d'un équipement réseau (routeur, switch ...). La loi de commande permet de distribuer temporellement le trafic qui traverse un nœud contrôlé dans les réseaux de communication filaires. Dans ce travail, nous avons considéré que les nœuds de communications sont des routeurs de type ALR. Pour traiter le problème de la consommation énergétique dans les réseaux de communication filaires, nous avons proposé un modèle énergétique ALR étendu adapté à la théorie de commande. Pour ce modèle, nous avons besoin de choisir deux paramètres (ß, ?), permettant de choisir la taille de file d'attente de référence qref et sa fenêtre temporelle d'actualisation Tqref .Ce deux paramètres ont été choisis à partir de plusieurs simulations avec différentes combinaisons des paramètres (ß, ?). Nous avons vu que la variation de ces deux paramètres permet d'agir énormément sur la QdS ainsi que sur la quantité d'énergie réduite. Les résultats théoriques sont ensuite testés sur Matlab-Simulink, puis sur le simulateur de réseaux NS-2. Les simulations ont montré que la consommation énergétique dans les réseaux de communication est bien réduite tout en garantissant un certain niveau de QdS.The computer tools (as the routers and calculators among others) present a high energy consumption. This problem has been already included in mobile networks. The question of energy is just beginning to be considered for "fixed" large-scale systems that reach nowadays high sizes. The objective of this thesis is to address the problem of energy consumption in wired communication networks: provide a certain level of quality of service (QoS) with respect to the packet lost, response speed and robustness with respect to different sampling periods while controlling power consumption of the system. The goal is to design a method from the theory of control, which guarantees these QoS. This technique is applied locally to a network equipment (router, switch ...) and the control law used to distribute temporally the traffic through a controlled node in the wired communications networks. In this work, we considere that the communication between nodes are performed by routers ALR type. In order to deal with energy reduction problem, we propose an extended ALR energy model adapted to control theory. For this model, we need to choose two parameters (ß, ?) allowing to choose the queue length reference, qref, and the related update time-window, Tqref. These parameters have been chosen after performing some simulations with different combinations of parameters (ß, ?). We have seen that the variation of these two parameters provide an impact over the QoS as well as the energy reduction. The theoretical results are then tested in Matlab-Simulink as well as some experiments under the simulator NS-2. Simulations showed that the energy consumption in communications networks is reduced while ensuring a certain level of QoS
Dispelling the Myths Behind First-author Citation Counts
Full text linkWe conducted a full-scale evaluative citation analysis study of scholars in the XML research field to explore just how different from each other author rankings resulting from different citation counting methods actually are, and to demonstrate the capability of emerging data and tools on the Web in supporting more realistic citation counting methods. Our results contest some common arguments for the continued
use of first-author citation counts in the evaluation of scholars, such as high correlations between author rankings by first-author citation counts and other citation
counting methods, and high costs of using more realistic citation counting methods that are not well-supported by the ISI databases. It is argued that increasingly available digital full text research papers make it possible for citation analysis studies to go beyond what the ISI databases have directly supported and to employ more
sophisticated methods
koamabayili/VECTRON-author-checklist: VECTRON author checklist
No full textWe have done our best to complete the author checklist relating to the use of animals in the hut study. Note that the objective for the hut study was to evaluate the IRS treatment applications for residual efficacy against Anopheles mosquitoes, including the local An. coluzzii mosquito population. Cows were only used to attract mosquitoes into the huts and no tests were carried out directly on the cows. The author checklist is intended for use with studies where experiments are carried out on animals, which is why we have had such difficulty in completing this for the hut study, as many of the questions do not relate to how the cows were used
- …
