Search

International audience; Texte de présentation de l'atelier animé dans le cadre du congrès et ayant permis l'expérimentation d'un serious game au format jeu de plateau.

International audience

International audience; This article is the second of a four-step symposium presenting the activity of the Junior Practices in Reflection Committee (JPR). It presents a bibliographical review and an example of a story by a young practitioner that take place within the framework of the JPRs. A design project is presented for a school registration reception in a municipality. It will present the methodology implemented with the project's actors, the tools used and its role in the project. All in order to consider the constraints of the agents and the public received and to ensure a quality of service. A professional storytelling, like the one presented, is used as a fulcrum for the discussion between young practitioner.

This thesis focuses mainly on two topics: one is the ionospheric signature of solar eclipses, the second is the Hole vs Enhancement debate about earthquake precursors. On the 21st August 2017 the shadow of a total eclipse drastically changed the state of the ionosphere over the USA. This effect is visible in the total electron content (TEC) measured by ~3000 GNSS stations seeing multiple GPS and GLONASS satellites. This tremendous dataset allows high-resolution characterization of the frequency content and wavelengths -using an omega-k analysis based on 3D Fast-Fourier-Transform (FFT)- of the eclipse signature in the ionosphere in order to fully identify traveling ionospheric disturbances (TIDs). We confirm the generation of TIDs associated with the eclipse including TIDs interpreted as bow waves in previous studies. Additionally we reveal, for the first time, short (50-100 km) and long (500-600 km) wavelength TIDs with periods between 30 and 65 min (Eisenbeis et al., 2019). On 2nd July 2019 another total solar eclipse happened across the South American continent at magnetic conjugate latitudes as the Great American Eclipse, and consequently useful to visualize the difference response. Although for the South American eclipse we have only data from more than hundred GNSS stations and located in a zone close to the sunset, we can show the clear evidence of the ionospheric signature of the eclipse (Eisenbeis & Occhipinti in prep.a).The second major topic in this work is the still ongoing debate about the possibility of earthquake precursors. Heki (2011) sparked this debate when he published results of the Tohoku earthquake showing a TEC enhancement before the earthquake. The enhancement claimed by Heki (2011) has been interpreted as a decrease in the background TEC after the seismic event, the so called ionospheric hole in literature. The existence of the enhancement has been promoted by several papers (e.g. He & Heki, 2017) extending the observation to several events with moderate magnitude (M> 7.5) and proposes a new vision of the rupture dynamics. By trying to reproduce their results we show that the reference curve used by Heki (2011) to define the TEC background is strongly affected by the order of polynomial fit as well as the selected time windows. This shows that the TEC enhancement could be, in fact, just an artifact, subjectively selected to create the presumed precursor (Eisenbeis & Occhipinti in prep.b).; Cette thèse porte principalement sur deux sujets: l'un est la signature ionosphérique des éclipses solaires, l'autre est le débat ‘Hole vs Enhancement’. Le 21 août 2017, l'ombre d'une éclipse totale a changé radicalement l'état de l'ionosphère au-dessus des Etats-Unis. Cet effet est visible dans le contenu total en électrons (TEC) mesuré par ~3000 stations GNSS qui voient des multiples satellites GPS et GLONASS. Ce formidable ensemble de données permet une caractérisation à haute résolution du contenu en fréquences et des longueurs d'onde - en utilisant une analyse omega-k basée sur la Transformée de Fourier Rapide (FFT) 3D - de la signature de l'éclipse dans l'ionosphère afin d'identifier complètement les perturbations ionosphériques mobiles (TID). Nous confirmons la génération de TIDs associées à l'éclipse, y compris les TIDs interprétées comme des ondes de proue dans les études précédentes. De plus, nous révélons, pour la première fois, des TID de courte (50-100 km) et de longue (500-600 km) longueurs d'onde avec des périodes entre 30 et 65 min (Eisenbeis et al., 2019). Le 2 juillet 2019, une autre éclipse solaire totale s'est produite à travers le continent sud-américain. Bien que nous n'ayons que des données provenant de plus de cent stations GNSS et situées dans une zone proche du coucher du soleil, nous pouvons montrer la preuve évidente de la signature ionosphérique de l'éclipse (Eisenbeis & Occhipinti in prep.a). Le deuxième grand sujet de ce travail est le débat sur la possibilité de précurseurs de séismes. Heki (2011) a suscité ce débat en publiant les résultats du séisme de Tohoku montrant une amélioration de la TEC avant le séisme. L'amélioration revendiquée par Heki (2011) a été interprétée comme une diminution de la TEC après l'événement, le soi-disant trou ionosphérique dans la littérature. L'existence de l'amélioration a été promue par plusieurs articles (e.g. He & Heki, 2017) étendant l'observation à plusieurs événements de magnitude modérée (M> 7.5) et propose une nouvelle vision de la dynamique de rupture. En essayant de reproduire leurs résultats, nous montrons que la courbe de référence utilisée par Heki (2011) est affectée par l'ordre d'ajustement polynomial ainsi que par les fenêtres temporelles sélectionnées. Ceci montre que l'amélioration du TEC pourrait en fait n'être qu'un artefact, subjectivement sélectionné pour créer le précurseur présumé (Eisenbeis & Occhipinti in prep.b).

International audience; Drinking water supply networks play an essential role in protecting the human and economic wellbeing of the territories they serve. To ensure continued quality of service, organisations involved in water infrastructure asset management (WIAM) need to deal with a number of issues related to global change. This paper presents the results of an original interdisciplinary foresight approach carried out by a group of engineering and social scientists, in partnership with a number of stakeholders. The purpose was to examine various possible pathways for the future of a French territory. The full title of our foresight study is: 'Supplying water destined for human consumption in Nouvelle-Aquitaine (France) up until 2070'. Four scenarios, as contrasted as possible, were designed based on five components: organisation and operation of the water supply service, social demands in terms of drinking water, the governance context, territorial dynamics, and the overall context. We then ran further simulations to visualise what a given infrastructure network would look like under each set of assumptions, and under different territorial configurations. One significant advantage of our foresight approach is the educational value it has for stakeholders and water managers. Foresight makes the future potentially visible and provides an opportunity to discuss it, in order to able to inform decision-making. Key words | global change, participative foresight approach, performance of the drinking water service, water infrastructures asset management (WIAM) HIGHLIGHTS • We present an original interdisciplinary foresight approach. • Our topic is: 'Supplying water destined for human consumption in Nouvelle-Aquitaine (France) up until 2070'. • Four scenarios, as contrasted as possible, were designed and translated into possible developments and requirements in the infrastructure. • The aim is to support policy makers and WIAM organisations in assessing their preparedness in the face of global change.

International audience; On the 21 August 2017 the eclipse shadow drastically changed the state of the ionosphere over the United States. This effect on the ionosphere is visible in the total electron content measured by Global Navigation Satellite Systems (GNSS). The shadow moved with the supersonic speed of ~1,000 m/s over Oregon to ~650 m/s over South Carolina. In order to exhaustively explore the ionospheric signature of the eclipse, we use data of total electron content from ~3,000 GNSS stations seeing multiple Global Positioning System (GPS) and Global Navigation Satellite System (GLONASS) satellites to visualize the phenomena. This tremendous dataset allows high‐resolution characterization of the frequency content and wavelengths—using an omega‐k analysis based on 3‐D fast Fourier transform—of the eclipse signature in the ionosphere in order to fully identify traveling ionospheric disturbances (TIDs). We confirm the generation of TIDs associated with the eclipse including TIDs interpreted as bow waves in previous studies. Additionally, we reveal, for the first time, short (50–100 km) and long (500–600 km) wavelength TIDs with periods between 30 and 65 min. The sources of the revealed short wavelength TIDs are co‐located with the regions of stronger gradient of the EUV related to sunspots. Our work confirms and describes physical properties of the waves observable in the ionosphere during the Great American Eclipse.

International audience; Drinking water supply networks are essential for the economic activity of territories, for their attractiveness and for the well-being of inhabitants. Maintaining a high performance of service over time requires taking into consideration many variables such as ageing infrastructures, demographic changes, socioeconomic dynamics and climate change consequences. In this context, Water Infrastructures Asset Management (WIAM) faces several challenges. This article presents the results of an original forecasting approach based on an interdisciplinary research carried out by a group of engineering and social scientists, and stakeholders. The main objective was to design various possible pathways for the future of a French territory and craft scenarios for WIAM strategies to 2070. The article describes first the main issues WIAM is facing in a global change context. Second we present the original forecasting method used. Third we expose the results and particularly the four prospective scenarios that describe the possible pathways of WIAM by 2070. Fourth scenarios are applied to four case studies in order to build prospective infrastructure needs scenarios. And finally we propose their implications for WIAM strategies and discuss the benefits and limits of such an approach to improve water management.

Le lundi 4 décembre 2017, participant à une « Assemblée nationale pour les CHSCT », 160 militants pour la santé au travail, syndicalistes et professionnels, se retrouvaient pour dire leur colère de la suppression des CHSCT sous couvert d'une fusion des instances représentatives du personnel. Depuis, la loi du 29 mars 2018 (loi de ratification des 6 ordonnances dites « Travail ») a confirmé la mesure parmi d'autres, donnant en outre la possibilité d'accélérer le processus (art. 5) si les mandats actuels n'arrivent à échéance qu'en 2019. Les Comités d'Hygiène, Sécurité, Conditions de travail, n'existeront bientôt plus dans le privé, sauf catégories d'entreprises particulières. Plus vite encore que nous ne l'attendions, leur disparition est aussi à l'ordre du jour dans le secteur public.

Le syndicalisme, et la FSU en particulier, n'échappe évidemment pas à la numérisation. Or, autant on analyse (un peu) et dénonce la main mise des grands monopoles que sont les "GAFAM" (Google, Apple, Facebook, Amazon, Microsoft), autant l'utilisation de ces outils dans l'activité syndicale n'est pas interrogée. On est au contraire au coeur de multiples injonctions paradoxales.

International audience; Communication latency problems are universal and have become a major performance bottleneck as we scale in distributed computing architectures. Specifically, research institutes around the world have built specialized supercomputers with powerful computation units in order to accelerate scientific computation. However, the problem often comes from the communication side instead of the computation side. In this paper we first demonstrate the severity of communication latency problems. Then we use Lattice Quantum Chromo Dynamic (LQCD) simulations as a case study to show how value prediction techniques can reduce the communication overheads, thus leading to higher performance without adding more expensive hardware. In detail, we first implement a software value predictor on LQCD simulations: our results indicate that 22.15% of the predictions result in performance gain and only 2.65% of the predictions lead to rollbacks. Next we explore the hardware value predictor design, which results in a 20-fold reduction of the prediction latency. In addition, based on the observation that the full range of floating point accuracy may not be always needed, we propose and implement an initial design of the tolerance value predictor: as the tolerance range increases, the prediction accuracy also increases dramatically.

Ce texte est un plaidoyer pour que la question du numérique dans la société devienne un débat social, syndical, politique. Ceci nécessite de reconnaître que les enjeux liés au numérique ne sont pas seulement techniques, même s'ils le sont aussi, ce qui demande que chacun.e s'approprie les savoirs scientifiques de base pour être capable de poser des questions légitimes. Sans quoi, le numérique risque de devenir comme la finance : de même que nous parlons de « la main invisible du marché », nous pourrions être soumis à notre insu à cette « main invisible du numérique.

International audience; Cloud computing has evolved to provide computing resources on-demand through a virtualized infrastructure, letting applications, computing power, data storage, and network resources to be provisioned and managed over private networks or over the Internet. Cloud services normally run on large data centers and demand a huge amount of electricity. Consequently, the electricity cost represents one of the major concerns of data centers, because it is sometimes nonlinear with the capacity of the data centers, and it is also associated with a high amount of carbon emission (CO2). However, energy-saving schemes that result in too much degradation of the system performance or in violations of service-level agreement (SLA) parameters would eventually cause the users to move to another cloud provider. Thus, there is a need to reach a balance between energy savings and the costs incurred by these savings in the execution of the applications. Therefore, in this paper, we propose and evaluate a power and SLA-aware application consolidation solution for cloud federations. It comprises a multi-agent system for server consolidation, taking into account SLA, power consumption, and carbon footprint. Different for similar solutions available in the literature, in our solution, when a cloud is overloaded, its data center needs to negotiate with other data centers before migrating the workload to another cloud. Simulation results show that our approach can reduce up to 46% of the power consumption while trying to meet performance requirements. Furthermore, we show that federated clouds can provide an adequate solution to deal with power consumption in the clouds.

International audience; Nowadays, cloud users face three important problems: (a) choosing one or more appropriate cloud provider(s) to run their application(s), (b) selecting appropriate cloud resources, which implies having enough information about the available resources, including their characteristics and constraints, and (c) configuring the cloud resources. These problems are mostly due to the wide range of resources. These resources usually have distinct dependencies, and they are offered at various clouds' layers. In this complex scenario, the users often have to handle cloud resources and their dependencies manually. This is an error-prone and time-consuming activity, even for skilled cloud users and system administrators. In this context, this paper proposes a software product line engineering (SPLE) method and a tool to deal with these issues. Our SPL-based engineering method enables a declarative and goal-oriented strategy. Furthermore, it allows resource selection and configuration in inter-cloud environments. In our proposal, the cloud users specify their applications and requirements, and our tool automatically selects and configures a suitable computing environment, taking into account temporal and functional dependencies. Experimental results on Amazon EC2 and Google Compute Engine (GCE) show that our approach enables unskilled users to have access to advanced inter-cloud computing configurations, without being concerned with the characteristics of each cloud.

International audience

International audience; This article explores the role of teacher education in foreign languages and cultures on the implementation of pedagogical practices aiming at developing learner's autonomy. During the " Master de Français Langue Etrangère et de Didactique de l'Anglais " at Lille 3 University, teacher trainees have a guided self-learning course in a foreign language. This self-learning allows them to practically experience theoretical notions related to the learner autonomy pedagogical paradigm, notions they tackled during other courses of their curriculum from a theoretical point of view. For our purposes, we interviewed some former students which are now foreign language teachers. Our main aim is to know if their experience of self-learning has some links on their actual pedagogical practices, especially concerning practices aiming at developing learners' autonomy.; Le présent article interroge le rôle de la formation de formateurs en didactique des langues étrangères dans la mise en place par les futurs formateurs en langues de pratiques à visée autonomisante. Dans le cadre des Master de Français Langue Etrangère et de Didactique de l'Anglais de l'Université Lille 3, les apprentis enseignants et formateurs font l'expérience d'une autoformation guidée dans une langue étrangère. Cette autoformation permet aux apprentis enseignants et formateurs de faire l'expérience pratique de notions théoriques relatives au paradigme pédagogique de l'autonomie de l'apprenant abordées dans d'autres cours de la formation. A travers des interviews avec des anciens étudiants de cette formation devenus maintenant des formateurs en langues, notre étude vise à comprendre les éventuels liens entre cette expérience d'autoformation et la mise en place de pratiques pédagogiques autonomisantes.

Carbon metabolism and virulence determinant production are often linked in pathogenic bacteria, and several regulatory elements have been reported to mediate this linkage in Staphylococcus aureus. Previously, we described a novel protein, catabolite control protein E (CcpE) that functions as a regulator of the tricarboxylic acid cycle. Here we demonstrate that CcpE also regulates virulence determinant biosynthesis and pathogenesis. Specifically, deletion of ccpE in S. aureus strain Newman revealed that CcpE affects transcription of virulence factors such as capA, the first gene in the capsule biosynthetic operon; hla, encoding α-toxin; and psmα, encoding the phenol-soluble modulin cluster α. Electrophoretic mobility shift assays demonstrated that CcpE binds to the hla promoter. Mice challenged with S. aureus strain Newman or its isogenic ΔccpE derivative revealed increased disease severity in the ΔccpE mutant using two animal models; an acute lung infection model and a skin infection model. Complementation of the mutant with the ccpE wild-type allele restored all phenotypes, demonstrating that CcpE is negative regulator of virulence in S. aureus.

International audience; Power consumption has became a critical concern in modern computing systems for various reasons including financial savings and environmental protection. With battery powered de-vices, we need to care about the available amount of energy since it is limited. For the case of supercomputers, as they imply a large aggregation of heavy CPU activities, we are exposed to a risk of overheating. As the design of current and future hardware is becoming more and more complex, energy prediction or estimation is as elusive as that of time performance. However, having a good prediction of power consumption is still an important request to the computer science community. Indeed, power consumption might become a common performance and cost metric in the near future. A good methodology for energy prediction could have a great impact on power-aware programming, compilation, or runtime monitoring. In this paper, we try to understand from measurements where and how power is consumed at the level of a computing node. We focus on a set of basic programming instructions, more precisely those related to CPU and memory. We propose an analytical prediction model based on the hypothesis that each basic instruction has an average energy cost that can be estimated on a given architecture through a series of micro-benchmarks. The considered energy cost per operation includes both the overhead of the embedding loop and associated (hardware/software) optimizations. Using these precalculated values, we derive a linear extrapolation model to predict the energy of a given algorithm expressed by means of atomic instructions. We then use three selected appli-cations to check the accuracy of our prediction method by comparing our estimations with the corresponding measurements obtained using a multimeter. We show a 9.48% energy prediction on sorting.

International audience; IaaS providers often allow the users to specify many re-quirements for their applications. However, users without advanced technical knowledge usually do not provide a good specification of the cloud environment, leading to low per-formance and/or high monetary cost. In this context, the users face the challenges of how to scale cloud-unaware ap-plications without re-engineering them. Therefore, in this paper, we propose and evaluate a cloud architecture, namely Excalibur, to execute applications in the cloud. In our ar-chitecture, the users provide the applications and the archi-tecture sets up the whole environment and adjusts it at run-time accordingly. We executed a genomics workflow in our architecture, which was deployed in Amazon EC2. The ex-periments show that the proposed architecture dynamically scales this cloud-unaware application up to 10 instances, re-ducing the execution time by 73% and the cost by 84% when compared to the execution in the configuration specified by the user.

Power consumption has became a critical concern in modern computing systems for various reasons including financial savings and environmental protection. With battery powered devices, we need to care about the available amount of energy since it is limited. For the case of supercomputers, as they imply a large aggregation of heavy CPU activities, we are exposed to a risk of overheating. As the design of current and future hardware is becoming more and more complex, energy prediction or estimation is as elusive as that of time performance. However, having a good prediction of power consumption is still an important request to the computer science community. Indeed, power consumption might become a common performance and cost metric in the near future. A good methodology for energy prediction could have a great impact on power-aware programming, compilation, or runtime monitoring. In this paper, we try to understand from measurements where and how power is consumed at the level of a computing node. We focus on a set of basic programming instructions, more precisely those related to CPU and memory. We propose an analytical prediction model based on the hypothesis that each basic instruction has an average energy cost that can be estimated on a given architecture through a series of micro-benchmarks. The considered energy cost per operation includes all of the overhead due to context of the loop where it is executed. Using these precalculated values, we derive an linear extrapolation model to predict the energy of a given algorithm expressed by means of atomic instructions. We then use three selected applications to check the accuracy of our prediction method by comparing our estimations with the corresponding measurements obtained using a multimeter. We show a 9.48\% energy prediction on sorting.

Quantum Chromodynamics (QCD) is the theory of strong nuclear force, responsible of the interactions between sub-nuclear particles. QCD simulations are typically performed through the lattice gauge theory approach, which provides a discrete analytical formalism called LQCD (Lattice Quantum Chromodynamics). LQCD simulations usually involve generating and then processing data on petabyte scale which demands multiple teraflop-years on supercomputers. Large parts of both, generation and analysis, can be reduced to the inversion of an extremely large matrix, the so-called Wilson-Dirac operator. For this purpose, and because this matrix is always sparse and struc-tured, iterative methods are definitely considered. Therefore, the procedure of the application of this operator, resulting in a vector-matrix product, appears as a critical computation kernel that should be optimized as much as possible. Evaluating the Wilson-Dirac operator involves symmetric stencil calculations where each node has 8 neighbors. Such configuration is really hindering when it comes to memory accesses and data exchanges among processors. For current and future generation of supercomputers the hierarchical memory structure make it next to impossible for a physicist to write an efficient code. Addressing these issues in other to harvest an acceptable amount of computing cycles for the real need, which means reaching a good level of efficiency, is the main concern of this paper. We present here a Domain Specific Language and corresponding toolkit, called QIRAL, which is a complete solution from symbolic notation to simulation code.

International audience; As data centers increase in size and computational capac- ity, numerous infrastructure issues become critical. Energy efficient is one of these issues because of the constantly increasing power consump- tion of CPUs, memory, and storage devices. A study shows that the whole energy consumed by data centers will be extremely high and it is like to overtake airlines in terms of carbon emissions. In that scenario, Cloud computing is gaining popularity since it can help companies to reduce costs and carbon footprint, usually distributing execution of ser- vices across distributed data centers. The research aims of this work are to propose and evaluate a Model for Federated Clouds that takes into account power consumption and Quality of Service (QoS) requirements. In our model, the energy reduction shall not result in negative impacts to the agreements between Cloud users and Cloud providers. Therefore, the model should ensure both energy-efficiency and QoS parameters, which sets up possibly conflicting objectives.

National audience; Enfin un véritable manuel d'informatique pour les lycéens et leurs professeurs !Les quatre concepts de machine, d'information, d'algorithme et de langage sont au coeur de l'informatique, et l'objet de ce cours est de montrer comment ils fonctionnent ensemble. En première partie, nous apprendrons à écrire des programmes, en découvrant les ingrédients qui les constituent : l'affectation, la séquence et le test, les boucles, les types, les fonctions et les fonctions récursives. Dans la deuxième partie, on verra comment représenter les informations que l'on veut communiquer, les stocker et les transformer - textes, nombres, images et sons. On apprendra également à structurer et compresser de grandes quantités d'informations, à les protéger par le chiffrement. On verra ensuite que derrière les informations, il y a toujours des objets matériels : ordinateurs, réseaux, robots, etc., qui font partie de notre quotidien. Enfin, on s'initiera à des savoir-faire utiles au XXIe siècle : ajouter des nombres exprimés en base deux, dessiner, retrouver une information par dichotomie, trier des informations et parcourir des graphes.Ce cours comporte des chapitres élémentaires et avancés. Chacun contient une partie de cours, des sections de savoir-faire qui permettent d'acquérir les capacités essentielles, et des exercices, notés difficiles pour certains, avec corrigé lorsque nécessaire.A qui s'adresse ce livre ?Ce manuel de cours est destiné aux élèves de terminale ayant choisi la spécialité Informatique et sciences du numérique au lycée ; il s'appuie sur le langage de programmation Python (version 3). Il sera également lu avec profit par tous les professionnels de l'informatique, qu'ils soient autodidactes ou non.

LPT-Orsay-14-142; International audience; We present here our ongoing work on a Domain Specific Language which aims to simplify Monte-Carlo simulations and measurements in the domain of Lattice Quantum Chromodynamics. The tool-chain, called Qiral, is used to produce high-performance OpenMP C code from LaTeX sources. We discuss conceptual issues and details of implementation and optimization. The comparison of the performance of the generated code to the well-established simulation software is also made.

National audience; Cet article présente la mise en œuvre et les principes d'un dispositif d'autoformation guidée en langues proposé dans le cadre d'un Master de Didactique de Langues Vivantes à l'Université de Lille3. Suite à une enquête auprès d'anciens étudiants de ce Master devenus formateurs en langues, nous avons analysé les liens entre cette expérience d'autoformation et la mobilisation des mêmes principes dans différents contextes. Notre enquête montre que la mise en place de pratiques autonomisantes s'accompagne d'un repositionnement de l'enseignant.

National audience; Manuel scolaire de la spécialité ISN en Terminale S, l'objectif du livre est d'introduire et donner une ensemble maximal d'éléments pour permettre cet enseignement nouveau, orienté autour de projets, en tenant compte de multiples variantes selon les axes choisis. Il est donc un ouvrage de référence, des savoirs et savoirs faire fondamentaux, mais aussi des contenus optionnels qui les entourent. Il s'articulent autour des quatre concepts de l'Informatique, mais surtout de montrer la manière dont ils fonctionnent ensemble. Quand nous étudierons les algorithmes fondamentaux, nous les exprimerons souvent dans un langage de programmation. Quand nous étudierons l'organisation des machines, nous verrons comment elles permettent d'exécuter des programmes exprimés dans un langage de programmation. Quand nous étudierons la notion d'information, nous verrons des algorithmes de compression, de chiffrement, etc. Ce livre est donc organisé en quatre parties regroupé en vingt-deux chapitres.

International audience; Reversible computing aims at keeping all information on input and intermediate values available at any step of the computation, making information virtually present everywhere. Rematerialization in register allocation amounts to recomputing values instead of spilling them in memory when registers run out. In this paper we detail a heuristic algorithm for exploiting reverse computing for register materialization. This improves information locality as it provides more opportunities for retrieving data. Rematerialization adds instructions and we show on one specifically designed example that reverse computing may alleviate the impact of these additional instructions on performance. We also show how thread parallelism may be optimized on GPUs by performing register allocation with reverse recomputing that increases the number of threads per Streaming Multiprocessor. This is done on the main kernel of Lattice Quantum Chromo Dynamics simulation program where we gain a 11 % speedup.

International audience; no abstract

Reversible computing could be in more or less long term mandatory for minimizing heat dissipation inherent to computing. It aims at keeping all information on input and intermediate values available at any step of the computation. Rematerialization in register allocation amounts to recomputing values instead of spilling them in memory when registers run out. In this paper we detail a heuristic algorithm for performing reverse register materialization and we use the high memory demanding LQCD (Lattice Quantum ChromoDynamics) application to demonstrate that important gains of up to 33% on register pressure can be obtained. This in turn enables an increase in Instruction-Level Parallelism and Thread-Level Parallelism. We demonstrate a 16.8% (statically timed) gain over a basic LQCD computation. Basic ideas of the algorithm and experimental results were already presented in a poster of another conference.

International audience

International audience; In this paper we address the issue of making a program reversible in terms of spatial complexity. Spatial complexity is the amount of memory/register locations required for performing the computation in both forward and backward directions. Spatial complexity has important relationship with the intrinsics power consumption required at run time; this was our primary motivation. But it has also important relationship with the trade off between storing or recomputing reused intermediate values, also known as the rematerialization problem in the context of compiler register allocation, or the checkpointing issue in the general case. We present a lower bound of the spatial complexity of a DAG (directed acyclic graph) with reversible operations, as well as a heuristic aimed at finding the minimum number of registers required for a forward and backward execution of a DAG . We define energetic garbage as the additional number of registers needed for the reversible computation with respect to the original computation. We have run experiments that suggest that the garbage size is never more than 50% of the DAG size for DAGs with unary/binary operations.

International audience; The current trend in electronics is to integrate more and more transistors on a chip and produce massive hardware resources. As a consequence, traditional computing models, which mainly compute in the temporal domain, do not work well anymore since it becomes increasingly difficult to orchestrate these massive-scale hardware resources in a centralized way. Spatial computing is a unifying term that embodies many unconventional computing models and means computing on a relatively homogeneous physical medium made of hardware components, where the communication time is dependent on the euclidean distance between the components (locality constraint). This constraint makes the programming for high performance significantly more complex compared to classical non-spatial hardware because performance now depends on where computation happens in space (mapping problem). Blob computing is a new approach that addresses this parallel computing challenge in a radically new and unconventional way: it decouples the mapping of computations onto the hardware from the software programming while still elegantly exploiting the space of the underlying hardware. Hardware mapping of computations is done by a physical force-based approach that simulates forces between threads of computation (automata). Attractive forces are used to keep automata that need to communicate with each other closer while repulsive forces are used for load balancing. The advantage of theses primitives is that they are simple enough to be implemented on an arbitrary computing medium. They form the basis of a runtime system (RTS) that transforms an arbitrary computing medium into an easier-to-program virtual machine called the blob machine. The basic objects of the blob machine are those automata, and the instructions let automata create new automata in specific ways so as to maintain a hierarchical organisation (which facilitates both the mapping and the programming). We detail the basic instructions of the blob machine and demonstrate their confluence. Programming a spatial medium to perform a given algorithm then boils down to programming the blob machine, provided the RTS is implemented on it. The advantage of this approach is the hardware independency, meaning that the same program can be used on different media. By means of several examples programmed using a high level langage description, we further show that we can efficiently implement most current parallel computing models, such as Single Instruction Multiple Data (SIMD), data parallelism, ''divide and conquer'' parallelism and pipelining which demonstrates parallel expressiveness. On sorting and matrix multiplication algorithms, we also show that our approach scales up optimally with the number of basic hardware components.

Technology can now produce massive hardware resources, large enough so that it becomes increasingly difficult to organize it in a centralized way. Spatial computing proposes to model such huge hardware as a relatively homogeneous computing medium satisfying a locality constraint: communication time is related to geometric distance. While the constraint is weak enough to allow arbitrary scalability, it is strong enough to make the tasks of programming and mapping significantly more complex; programming thus becomes the central problem. We propose a two level programming approach: at low level, a run time system layer runs on the computing medium and transforms it into a virtual machine, called the blob machine; at high level, programs run on this more expressive virtual machine. The system layer is implemented on the computing medium as a local rule that manages distributed objects similar to membranes and filament channels; system calls are distributed primitives that allow objects to be created or deleted. The physical interpretation of the objects allows the implementation on arbitrary spatial computing medium. This layer is responsible for placing the objects in the space and does it by simulating physical forces. Each object is controlled by a FSA (Finite State Automaton) whose output actions are the object primitives (creation \& deletion). The set of FSAs together with the communication pathways implied by the topology of encapsulated membranes and channels, define a network of FSA that can self develop. This ``self-developing network of FSA'' is the formal definition of the blob machine. We present this blob machine, and how to program it using a higher level language description. We illustrate the execution of many examples of small and simple programs that cover a wide spectrum of parallel paradigms, including SIMD, data parallelism, Divide and Conquer and pipelining. Usually, if the problem needs a space of O(n), the time complexity can be reduced to O(n^{1/d}) where d is the dimensionality of the computing medium: 2D, or 3D. In some cases the same program runs optimally for any value of d, in other cases, the dimensionality must be considered. Citation about a computer using physical objects as primitives, like blobs and channels, which has given us inspiration: `The kinematic model [of self reproducing machine] deals with the geometric-kinematic problems of movement, contact, positioning, fusing and cutting'', John Von Neumann, Theory of Self-Reproducing Automata, 1966.

Dans le cadre de l'enseignement / apprentissage des langues en Master de didactique des langues à l'université de Lille 3 sont proposés depuis deux ans des dispositifs d'autoformation guidée pour l'anglais, l'espagnol et le FLE. Ces dispositifs hybrides fonctionnent autour de plusieurs pôles : un Centre de Ressources en Langues où sont classées des ressources multimédia, des regroupements sous forme d'ateliers "apprendre à apprendre" et d'expression orale, une plateforme, lieu d'échanges et de travail collaboratif. Parallèlement, des entretiens sont proposés aux étudiants à qui il est demandé de rédiger un journal d'apprentissage.Notre corpus est constitué de questionnaires remplis par les étudiants ayant participé au dispositif, croisés avec les carnets de bord rédigés au long de la formation. L'analyse des données semble aboutir à l'idée que l'autonomie ne se développe pas dans toutes les activités de la même manière et surtout que l'on voit surgir différentes formes d'autonomie (autonomie générale, langagière et d'apprentissage). Comment s'articulent ces éléments constitutifs de l'autonomie ; la place qu'ils tiennent dans le dispositif est-elle fixe ou évolue-t-elle selon des étapes prévisibles ? Nous avons essayé à partir de ces éléments de faire surgir les tensions qui apparaissent entre guidage et autonomie. Ces notions semblent contradictoires en apparence, mais il apparaît que certaines formes de guidage peuvent favoriser en fin de compte l'autonomie. Doit-on considérer qu'il s'agit alors d'un passage nécessaire pour permettre ensuite de développer des stratégies individuelles adaptées ? Notre objectif est de parvenir à comprendre selon quelle articulation ces éléments doivent être mis en place afin d'optimiser le dispositif d'autoformation. A guided self-access environment in English, French as a foreign language and Spanish is open to students in the "Didactics of Foreign Languages" Master's at Lille3 University. This blended environment includes the following elements: a Language Resource Centre with multimedia resources that have been selected and classified according to specific criteria, group sessions such as Learning to Learn workshops or Speaking Practice sessions and a Moodle collaborative e-learning platform. Individual interviews are organised for the students who also have to maintain a logbook of their learning process.Our corpus consists of the answers to a questionnaire completed by students who took part in the blended learning, cross-checked with their logbooks for the semester. An analysis of this corpus indicates that autonomy does not develop equally from the various activities and moreover that different forms of autonomy (general autonomy, language autonomy and learning autonomy) emerge during the learning process. How do these different aspects of autonomy combine? Are they established once and for all, or do they evolve, and if so, how? Are there any predictable steps? Our aim is to understand the tensions that emerge between guidance and autonomy, two notions that may seem contradictory, but which, in the final analysis, appear to be complementary in that certain forms of guidance favour the development of autonomy. Should we consider guidelines as necessary and useful, enabling students to develop appropriate individual strategies? From this analysis, our goal is to determine the essential elements of the learning environment, and the articulation between them, in order to optimise its efficiency for successful, autonomous language learning.

This is a position paper introducing blob computing: A Blob is a generic primitive used to structure a uniform computing substrate into an easier-to-program parallel virtual machine. We find inherent limitations in the main trend of today's parallel computing, and propose an alternative unifiying model trying to combine both scalability and programmability. We seek to program a uniform computing medium such as fine grain 2D cellular automata, or more generally coarse grain 2D grids of Processing Elements, using two levels: In the first "system level", a local rule or run time system is implemented on the computing medium. A blob is similar to a deformable elastic membrane filled with a gas of atoms. Blobs are interconnected using channels, which act as a spring to bring connected blobs closer to each other. The system implements in a distributed way: movement, duplication and deletion of blobs and channels. It can also propagate waves to communicate signals intra-blob, or inter-blob. In the second "programmable level", each blob and channel contains a finite state automaton, with output instruction triggering duplication or deletion . Execution starts with a single ancestor blob that duplicates and creates channels repeatedly, thus generating a network of automata. It installs a higher level virtual machine on top of a low level uniform computing medium. This "blob machine" is an example of "self developing automata network". This work is inspired by the Von Neumann's remarks on "the kinematic model [of self reproducing machine]".

%HEVEA\publinkCDEPPP06.ps.gz; International audience; no abstract

This article reviews the setting up of a multimedia workshop which took place within the context of French as a Foreign Language at University and in which the students had to design and build their own Web Site. The discussion is based on interviews with some students who participated in the project. We describe the project and its analysis according to three main lines of inquiry: the specificities of the Internet, promoting motivation, collaborative learning and acquisition of specific vocabulary and abilities; the new kinds of human mediation, altering the roles of both student and teacher and providing a variety of interactions in the different tasks; the operative and autonomizing perspective of such a workshop. The conclusion offers some directions to repeat or expand this project for a better pedagogical and institutional integration., Cet article fait le point sur la mise en place d'un atelier de création d'un site Internet dans le cadre de l'enseignement universitaire du Français Langue Étrangère, réflexion fondée entre autres sur un corpus d'entretiens avec des étudiants ayant participé à ce projet. Ce bilan s'articule autour de trois axes de réflexion : les spécificités du support Internet, source de motivation et base d'un apprentissage collaboratif et de l'acquisition d'un lexique et d'un savoir-faire spécifiques ; la place de la médiation humaine, la redistribution des rôles et la diversité des interactions dans les différentes tâches mises en œuvre ; la dimension actionnelle et autonomisante d'un tel projet. La conclusion propose des pistes pour reconduire ou prolonger cet atelier en vue d'une meilleure intégration pédagogique et institutionnelle.

Cet article fait partie d'une sélection d'articles faisant suite au colloque Untele, "L'autonomie de l'enseignant et de l'apprenant face aux Technologies de l'Information et de la Communication". Colloque Organisé à l'Université de Technologie de Compiègne en mars 2004.; This article reviews the setting up of a multimedia workshop which took place within the context of French as a Foreign Language at University and in which the students had to design and build their own Web Site. The discussion is based on interviews with some students who participated in the project. We describe the project and its analysis according to three main lines of inquiry: the specificities of the Internet, promoting motivation, collaborative learning and acquisition of specific vocabulary and abilities; the new kinds of human mediation, altering the roles of both student and teacher and providing a variety of interactions in the different tasks; the operative and autonomizing perspective of such a workshop. The conclusion offers some directions to repeat or expand this project for a better pedagogical and institutional integration.; Cet article fait le point sur la mise en place d'un atelier de création d'un site Internet dans le cadre de l'enseignement universitaire du Français Langue Étrangère, réflexion fondée entre autres sur un corpus d'entretiens avec des étudiants ayant participé à ce projet. Ce bilan s'articule autour de trois axes de réflexion : les spécificités du support Internet, source de motivation et base d'un apprentissage collaboratif et de l'acquisition d'un lexique et d'un savoir-faire spécifiques ; la place de la médiation humaine, la redistribution des rôles et la diversité des interactions dans les différentes tâches mises en œuvre ; la dimension actionnelle et autonomisante d'un tel projet. La conclusion propose des pistes pour reconduire ou prolonger cet atelier en vue d'une meilleure intégration pédagogique et institutionnelle.

%HEVEA\publinkCDEPPP05.ps.gz; International audience; We propose a programming model dedicated to real-time videostreamingapplications for embedded media devices, including highdefinitionTVs. This model is built on the synchronous programmingmodel extended with domain-specific knowledge — periodic evolutionof streams — to allow correct-by-construction properties of theapplication to be proven by the compiler. These properties includebuffer requirements and delays between input and output streams.Such properties are tedious to analyze by hand, due to the combinatoricsof video filters, multiple data rates and formats. We show howto extend a core synchronous data-flow language with a notion of periodicclocks, and to design a relaxed clock calculus (a type systemfor clocks) to allow non strictly synchronous processes to be composed.This relaxation is associated with a subtyping rule in the clockcalculus. Delay, buffer insertion and control code for these buffersare automatically inferred from the clock types through a systematicprogram transformation.

We propose a programming model and language dedicated to high-performance streaming applications. In particular, we study real-time video-streaming for embedded media devices, including high-definition TVs. This language builds on the synchronous programming model and on domain-specific knowledge --- periodic evolution of streams --- to allow correct-by-construction properties of the application to be proven by the compiler. These properties include buffer requirements and delays between input and output streams. Correctness of the implementation is difficult to assess with traditional (asynchronous) approaches. Such properties are tedious to analyze by hand, due to the combinatorics of video filters, multiple data rates and formats. For example, the design of communicating buffers between filtering processes (image scaling, quality enhancement, etc.) whose clocks do not strictly match is tedious and error-prone. Two communicating periodic processes are defined as n-synchronous if they can be implemented in the ordinary (0-)synchronous model with a FIFO buffer of size n. We extend a core synchronous data-flow language with a notion of periodic clocks, and design a relaxed clock calculus (a type system for clocks) to allow non strictly synchronous processes to be composed. This relaxation is associated with a subtyping rule in the clock calculus. Delay, buffer insertion and control code for these buffers are automatically inferred from the clock types through a systematic program transformation.

This article reviews the setting up of a multimedia workshop which took place within the context of French as a Foreign Language at University and in which the students had to design and build their own Web Site. The discussion is based on interviews with some students who participated in the project. We describe the project and its analysis according to three main lines of inquiry: the specificities of the Internet, promoting motivation, collaborative learning and acquisition of specific vocabulary and abilities; the new kinds of human mediation, altering the roles of both student and teacher and providing a variety of interactions in the different tasks; the operative and autonomizing perspective of such a workshop. The conclusion offers some directions to repeat or expand this project for a better pedagogical and institutional integration., Cet article fait le point sur la mise en place d'un atelier de création d'un site Internet dans le cadre de l'enseignement universitaire du Français Langue Étrangère, réflexion fondée entre autres sur un corpus d'entretiens avec des étudiants ayant participé à ce projet. Ce bilan s'articule autour de trois axes de réflexion : les spécificités du support Internet, source de motivation et base d'un apprentissage collaboratif et de l'acquisition d'un lexique et d'un savoir-faire spécifiques ; la place de la médiation humaine, la redistribution des rôles et la diversité des interactions dans les différentes tâches mises en œuvre ; la dimension actionnelle et autonomisante d'un tel projet. La conclusion propose des pistes pour reconduire ou prolonger cet atelier en vue d'une meilleure intégration pédagogique et institutionnelle.

International audience; Register allocation in loops is generally performed after or during the software pipelining process. This is because doing a conventional register allocation as a first step without assuming a schedule lacks the information of interferences between values live ranges. Thus, the register allocator may introduce an excessive amount of false dependences that dramatically reduce the ILP (Instruction Level Parallelism). We present a new theoretical framework for controlling the register pressure before software pipelining. This is based on inserting some anti-dependence edges (register reuse edges) labeled with reuse distances, directly on the data dependence graph. In this new graph, we are able to fix the register pressure, measured as the number of simultaneously alive variables in any schedule. The determination of register and distance reuse is parameterized by the desired minimum initiation interval (MII) as well as by the register pressure constraints - either can be minimized while the other one is fixed. After scheduling, register allocation is done on conventional register sets or on rotating register files. We give an optimal exact model, and an approximation that generalizes the Ning-Gao [22] buffer optimization method. We provide experimental results which show good improvement compared to [22]. Our theoretical model considers superscalar, VLIW and EPIC/IA64 processors.

This report aims to help the end-users to use the Technical tools proposed in Care-W. This tools are presented in (Eisenbeis et al, 2002a). For each type of tools, two type of information are described: the data and the way to collect them; the tools and proposals to choose the tool according to the service.

International audience; The register allocation in loops is generally performed after or during the software pipelining process. This is because doing a conventional register allocation at first step without assuming a schedule lacks the information of interferences between variable lifetime intervals. Thus, the register allocator may introduce an excessive amount of false dependences that reduce dramatically the ILP (Instruction Level Parallelism).We present a new framework for controlling the register pressure before software pipelining. This is based on inserting some anti-dependences edges (register reuse edges) labeled with reuse distances, directly on the data dependence graph. In this new graph, we are able to guarantee that the number of simultaneously alive variables in any schedule does not exceed a limit. The determination of register and distance reuse is parameterized by the desired critical circuit ratio (MII) as well as by the register pressure constraints - either can be minimized while the other one is fixed. After scheduling, register allocation is done cyclically on conventional register sets or on rotating register files. We give an optimal exact model, and another approximative one that generalizes the Ning-Gao [13] buffer optimization heuristics.

Cemagref is responsible for WP2, which is divided in three tasks. This report refers to the task 2.2. This task has several objectives: to test and compare the models on several water networks; to have a critical look on the models, with the aim of validating and fitting them; to improve and make their utilisation easier; to help to define a best use procedure.

The convergence of digital television, Internet access, gaming, and digital media capture and playback stresses the importance of high-quality and high-performance video and graphics processing. The SANDRA project, a collaboration between Philips Research and INRIA, develops a consistent and efficient system design approach for regular, real-time constrained stream processing. The project aims at providing a system template with its associated compiler chain and application development framework, enabling an early validation of both the functional and the non-functional requirements of the application at every system design stage.

Urban water networks represent considerable assets. Maintenance and rehabilitation of these buried infrastructures have become of the greatest importance for public or private managers. Water pipes, mostly in grey cast iron or ductile iron, are subjects to the combined effects of corrosion and loads. The impacts of the aging process concern the quality of service (interruptions, water quality, ..), operation and maintenance costs, as well as the safety of the urban environment. The setting up of an asset management policy (including corrective and preventive measures) relies on the definition and the evaluation of performance indicators and implies the development of large databases. One major use of these databases is the development of aging models in order to help managers in forecasting needs for investigations or rehabilitation and in defining priorities. These models are based on statistical analysis (survival analysis and/or regression techniques, etc.), and can be used to define pipe categories and corresponding failure rates and/or to predict the aging of the asset as a whole. Previous studies on few sites have shown the interest of these models. In addition, economical approaches are in progress. These models and approaches are going to be studied on test cities in the framework of a European research project., Les réseaux urbains d'eau potable constituent des patrimoines considérables dont la maintenance et la réhabilitation commencent à devenir une préoccupation majeure des gestionnaires publics ou privés. Les conduites, généralement en fonte, subissent les effets conjoints des charges mécaniques et des processus de corrosion. Les impacts de ce vieillissement concernent la qualité du service rendu (qualité de l'eau distribuée, continuité du service, etc.) les coûts, mais aussi la sécurité de l'environnement bâti. La mise en place d'une gestion patrimoniale de ces réseaux enterrés (associant mesures correctives et actions préventives) repose sur la définition et l'évaluation d'indicateurs de performance et sur la constitution, par les gestionnaires d'un patrimoine de données. L'une des utilisations majeures de ces bases de données est la construction de modèles de vieillissement utilisables pour prévoir et hiérarchiser les besoins (investigations et réhabilitations). Ces modèles sont basés sur des analyses statistiques (analyses de survie, modèles de régression, etc.). Ils permettent de classer les conduites par niveau de risque de défaillance et/ou de prévoir le vieillissement d'un parc de canalisation. Les résultats disponibles sur quelques sites permettent en outre de motiver la constitution de bases de données pour la maintenance. Des approches économiques complémentaires sont également en cours d'élaboration. Ces méthodes, encore peu utilisées actuellement, pourront être opérationnelles grâce à l'informatisation et la complétion de la collecte des données ainsi qu'à de prochaines études européennes qui les testeront sur des sites pilotes.

International audience; Pervasive video processing in future Ambient Intelligence environments sets new challenges in embedded system design. In particular, very high performance requirements have to be combined with the constraints of deeply embedded systems, frequently changing operating modes, and low-cost, high-volume production. By leveraging upon the key properties of the application domain, we devised a computation model, a hardware template, and a programming approach which provide a natural mapping from application requirements to a complete system solution. Our approach enables the direct exploitation of concurrency and regularity in achieving the combined challenge of adaptability, performance, and efficiency.

A specific situation of learning has been analysed, in reference to the traditional classroom, based on the notions of reparation and readjustement. The activity described relies on the use of an educational software for oral comprehension, based on television reports and forms to be filled in. This activity is organized in pairs. The different characteristics of repairing schemes in traditional language class and in authentic communication are briefly described. Then a comparison is made thanks to the analysis of the transcription of the conversations that have taken place around a specific aid and in a very different situation. This comparison is made taking into account the following questions: the organisation of the repairing scheme and the origin of reparations, the aim of such reparations (see Bange's bifocalisation), the form taken by the reparations: injunction or incentive? Positive or negative judgement?) In our second part, our aim is to focus on the specificities of the reparations in such a setting and to work out the way oral comprehension is considered. The aim is to relate the change in the type of aid and context with the reparing schemes used in language learning., Nous avons analysé, en nous référant au groupe-classe traditionnel, une situation particulière d'apprentissage en adoptant comme entrée la question des réparations et des réajustements: il s'agit d'une activité à partir d'un didacticiel de compréhension orale sur la base de reportages télévisés et de questionnaires à remplir. Les apprenants mênent cette activité en binôme. Nous rappelons et analysons brièvement les différentes caractéristiques du schéma réparateur dans la classe de langue en général et dans les conversations authentiques. Nous établissons ensuite une comparaison à partir de l'analyse de transcriptions des échanges qui ont donc lieu autour d'un support et dans un contexte très différents, comparaison qui se mène essentiellement sur la base des questions suivantes: l'organisation du schéma réparateur et l'initiative des réparations, l'objet des réparations (cf. notion de bifocalisation chez Bange), la forme des réparations (correction-injonction ou correction incitative? appréciation positive ou négative?) La deuxième partie du travail vise à repérer les particularités des réparations mises en place dans ce dispositif et à cerner les incidences sur la manière dont la compréhension orale est travaillée. Notre objectif est de mettre en relation la modification des supports et des contextes avec les schémas réparateurs dans l'apprentissage d'une langue.

The work presented in this paper is carried out within the European project CARE-W (Computer Aided REhabilitation of Water networks. Contract N° EVK1-CT-2000-00053 5th FP), which deals with the development and testing of models and tools for computer aided support of sustainable strategies for water supply network rehabilitation. Three complementary paths to the management of network rehabilitation are followed: the definition of a comprehensive set of performance indicators, the calculation and evaluation of long-term strategic planning, and the prioritisation of projects in the annual rehabilitation plan of a utility according to different objectives. Here, the latter component is presented. The selection of rehabilitation projects requires the consideration of several points of view. A survey among 12 European water utilities revealed the different objectives of rehabilitation and their relative importance, as well as the information that are used for selecting rehabilitation projects. From this inquiry 15 criteria for comparing and evaluating rehabilitation projects were constructed. The criteria are of different nature: they can be cost functions, quantification of observed deficiencies, evaluation of risks, or evaluation of the contribution of a pipe to a problem that occurs in a particular zone. The utilisation of criteria for comparing and prioritising rehab actions is demonstrated.

Water pipe ages and failure rates force water services to consider carefully, the issue of renewal of their water network. A thorough diagnosis of current and predicted network condition is required before key decisions can be made. Failure rate is one of the main condition indicators and hence, adequate records of break occurrence and accurate location are highly desirable. Ideally, Trench opening gives a visual access to the pipe and its environment. This paper presents a proposed format for a data input, using maximum available information recorded after a repair. This information supplies databases or Geographical Information Systems and can be used in the framework of network ageing models, aimed at elaborating tools for renewal decision support., L'âge des conduites d'eau potable et la fréquence des défaillances incitent les gestionnaires à s'interroger sur la problématique du renouvellement des réseaux d'eau potable. Le processus de décision nécessite de pouvoir diagnostiquer l'état du réseau et de pouvoir suivre son évolution dans le temps. Les défaillances sont des indicateurs de l'état du réseau, l'enregistrement de l'événement et sa localisation précise sur la conduite sont une première source de connaissance et de suivi. En outre, l'ouverture d'une tranchée du fait de la réparation permet un accès visuel à la conduite et à son environnement immédiat, informations inaccessibles en temps normal. La fiche de renseignement sur les défaillances présentée dans cette communication propose un mode de saisie d'un maximum d'informations accessibles et disponibles lors d'une intervention. Ces informations permettent d'alimenter des bases de données ou des systèmes d'information géographique spécialisés et peuvent également être intégrées dans des modèles d'évolution du vieillissement du réseau dans le but d'élaborer des outils d'aide à la décision du renouvellement.