Indice
Letteratura scientifica selezionata sul tema "Programmation parallèle et distribuée multi-Niveaux"
Cita una fonte nei formati APA, MLA, Chicago, Harvard e in molti altri stili
Consulta la lista di attuali articoli, libri, tesi, atti di convegni e altre fonti scientifiche attinenti al tema "Programmation parallèle et distribuée multi-Niveaux".
Accanto a ogni fonte nell'elenco di riferimenti c'è un pulsante "Aggiungi alla bibliografia". Premilo e genereremo automaticamente la citazione bibliografica dell'opera scelta nello stile citazionale di cui hai bisogno: APA, MLA, Harvard, Chicago, Vancouver ecc.
Puoi anche scaricare il testo completo della pubblicazione scientifica nel formato .pdf e leggere online l'abstract (il sommario) dell'opera se è presente nei metadati.
Tesi sul tema "Programmation parallèle et distribuée multi-Niveaux"
Moukir, Sara. "High performance analysis for road traffic control". Electronic Thesis or Diss., université Paris-Saclay, 2024. http://www.theses.fr/2024UPASG039.
Testo completoThe need to reduce travel times and energy consumption in urban road networks is critical for improving collective well-being and environmental sustainability. Since the 1950s, traffic modeling has been a central research focus. With the rapid evolution of computing capabilities in the 21st century, sophisticated digital simulations have emerged, accurately depicting road traffic complexities. Mobility simulations are essential for assessing emerging technologies like cooperative systems and dynamic GPS navigation without disrupting real traffic.As transport systems become more complex with real-time information, simulation models must adapt. Multi-agent simulations, which analyze individual behaviors within a dynamic environment, are particularly suited for this task. These simulations help understand and manage urban traffic by representing interactions between travelers and their environment.Simulating large populations of travelers in cities, potentially millions of individuals, has historically been computationally demanding. Advanced computer technologies allowing distributed calculations across multiple computers have opened new possibilities. However, many urban mobility simulators do not fully exploit these distributed architectures, limiting their ability to model complex scenarios involving many travelers and extensive networks.The main objective of this research is to improve the algorithmic and computational performance of mobility simulators. We aim to develop and validate generic and reproducible distribution models that can be adopted by various multi-agent mobility simulators. This approach seeks to overcome technical barriers and provide a solid foundation for analyzing complex transport systems in dynamic urban environments.Our research leverages the MATSim traffic simulator due to its flexibility and open structure. MATSim is widely recognized in the literature for multi-agent traffic simulation, making it an ideal candidate to test our generic methods.Our first contribution applies the "Unite and Conquer" (UC) approach to MATSim. This method accelerates simulation speed by leveraging modern computing architectures. The multiMATSim approach involves replicating several MATSim instances across multiple computing nodes with periodic communications. Each instance runs on a separate node, utilizing MATSim's native multithreading capabilities to enhance parallelism. Periodic synchronization ensures data consistency, while fault tolerance mechanisms allow the simulation to continue smoothly even if some instances fail. This approach efficiently uses diverse computational resources based on each node's specific capabilities.The second contribution explores artificial intelligence techniques to expedite the simulation process. Specifically, we use deep neural networks to predict MATSim simulation outcomes. Initially implemented on a single node, this proof-of-concept approach efficiently uses available CPU resources. Neural networks are trained on data from previous simulations to predict key metrics like travel times and congestion levels. The outputs are compared to MATSim results to assess accuracy. This approach is designed to scale, with future plans for distributed neural network training across multiple nodes.In summary, our contributions provide new algorithmic variants and explore integrating high-performance computing and AI into multi-agent traffic simulators. We aim to demonstrate the impact of these models and technologies on traffic simulation, addressing the challenges and limitations of their implementation. Our work highlights the benefits of emerging architectures and new algorithmic concepts for enhancing the robustness and performance of traffic simulators, presenting promising results
Allombert, Victor. "Abstraction fonctionnelle pour la programmation d’architecture multi-niveaux : formalisation et implantation". Thesis, Paris Est, 2017. http://www.theses.fr/2017PESC1016/document.
Testo completoFrom personal computers using an increasing number of cores, to supercomputers having millions of computing units, parallel architectures are the current standard. The high performance architectures are usually referenced to as hierarchical, as they are composed from clusters of multi-processors of multi-cores. Programming such architectures is known to be notoriously difficult. Writing parallel programs is, most of the time, difficult for both the algorithmic and the implementation phase. To answer those concerns, many structured models and languages were proposed in order to increase both expressiveness and efficiency. Among other models, Multi-BSP is a bridging model dedicated to hierarchical architecture that ensures efficiency, execution safety, scalability and cost prediction. It is an extension of the well known BSP model that handles flat architectures.In this thesis we introduce the Multi-ML language, which allows programming Multi-BSP algorithms “à la ML” and thus, guarantees the properties of the Multi-BSP model and the execution safety, thanks to a ML type system. To deal with the multi-level execution model of Multi-ML, we defined formal semantics which describe the valid evaluation of an expression. To ensure the execution safety of Multi-ML programs, we also propose a typing system that preserves replicated coherence. An abstract machine is defined to formally describe the evaluation of a Multi-ML program on a Multi-BSP architecture. An implementation of the language is available as a compilation toolchain. It is thus possible to generate an efficient parallel code from a program written in Multi-ML and execute it on any hierarchical machine
Kortas, Samuel. "Préconditionnement multi-niveaux et multi-domaines de solveurs volumes finis de haute précision : implémentation sur machine parallèle". Aix-Marseille 1, 1997. http://www.theses.fr/1997AIX11091.
Testo completoHugues, Maxime. "Un paradigme de programmation multi-niveaux pour le calcul numérique sur les machines post-petascales et exascales". Thesis, Lille 1, 2011. http://www.theses.fr/2011LIL10146/document.
Testo completoThe coming of post-petscale and exascale supercomputers offers the perspective to accelerate the solving of engineering problems and to highly complex modeling. However, these future systems challenge computer scientists to built such machines. Many issues must be faced such as fault-tolerance, energy consumption and the programming of these complex systems composed of billion cores.In this thesis, we have focused on the programming aspect and propose a multi-level programming paradigm composed of three levels. For the low level, a data parallel paradigm is proposed to program many-cores processors for its focus on data mapping and movements. We have implemented and evaluated the SpMV with various sparse matrix formats on GPU to illustrate this point. For the intermediate level, we propose a message passing paradigm in order to optimize inter-sockets and inter-nodes communications. For the high level, a graph description paradigm is proposed to program and manage the parallelism between nodes.With a dense matrix inversion method developed in YML, we underline the interest of graph for the Time-To-Solution reduction and for the support of asynchronous communications in a transparent way. The interest of graph is also demonstrated for I/O optimizations and for their direct support into the programming model. We finally conclude by analyzing a such proposition of programming paradigm for exascale machines and outlines the future work direction
Sifakis, Emmanuel. "Programmation efficace et sécurisé d'applications à mémoire partagée". Phd thesis, Université de Grenoble, 2013. http://tel.archives-ouvertes.fr/tel-00823054.
Testo completoGurhem, Jérôme. "Paradigmes de programmation répartie et parallèle utilisant des graphes de tâches pour supercalculateurs post-pétascale". Thesis, Lille, 2021. http://www.theses.fr/2021LILUI005.
Testo completoSince the middle of the 1990s, message passing libraries are the most used technology to implement parallel and distributed applications. However, they may not be a solution efficient enough on exascale machines since scalability issues will appear due to the increase in computing resources. Task-based programming models can be used, for example, to avoid collective communications along all the resources like reductions, broadcast or gather by transforming them into multiple operations on tasks. Then, these operations can be scheduled by the scheduler to place the data and computations in a way that optimize and reduce the data communications. The main objective of this thesis is to study what must be task-based programming for scientific applications and to propose a specification of such distributed and parallel programming, by experimenting for several simplified representations of important scientific applications for TOTAL, and classical dense and sparse linear methods.During the dissertation, several programming languages and paradigms are studied. Dense linear methods to solve linear systems, sequences of sparse matrix vector product and the Kirchhoff seismic pre-stack depth migration are studied and implemented as task-based applications. A taxonomy, based on several of these languages and paradigms is proposed.Software were developed using these programming models for each simplified application. As a result of these researches, a methodology for parallel task programming is proposed, optimizing data movements, in general, and for targeted scientific applications, in particular
Wu, Xinzhe. "Contribution à l’émergence de nouvelles méthodes parallèles et réparties intelligentes utilisant un paradigme de programmation multi-niveaux pour le calcul extrême". Thesis, Lille 1, 2019. http://www.theses.fr/2019LIL1I010/document.
Testo completoKrylov iterative methods are frequently used on High-Performance Computing (HPC) systems to solve the extremely large sparse linear systems and eigenvalue problems from science and engineering fields. With the increase of both number of computing units and the heterogeneity of supercomputers, time spent in the global communication and synchronization severely damage the parallel performance of iterative methods. Programming on supercomputers tends to become distributed and parallel. Algorithm development should consider the principles: 1) multi-granularity parallelism; 2) hierarchical memory; 3) minimization of global communication; 4) promotion of the asynchronicity; 5) proposition of multi-level scheduling strategies and manager engines to handle huge traffic and improve the fault tolerance. In response to these goals, we present a distributed and parallel multi-level programming paradigm for Krylov methods on HPC platforms. The first part of our work focuses on an implementation of a scalable matrix generator to create test matrices with customized eigenvalue for benchmarking iterative methods on supercomputers. In the second part, we aim to study the numerical and parallel performance of proposed distributed and parallel iterative method. Its implementation with a manager engine and runtime can handle the huge communication traffic, fault tolerance, and reusability. In the third part, an auto-tuning scheme is introduced for the smart selection of its parameters at runtime. Finally, we analyse the possibility to implement the distributed and parallel paradigm by a graph-based workflow runtime environment
Certner, Olivier. "Environnement de programmation, support à l'exécution et simulateur pour machines à grand nombre de cœurs". Phd thesis, Université Paris Sud - Paris XI, 2010. http://tel.archives-ouvertes.fr/tel-00826616.
Testo completo