Sylvain Contassot-Vivier

University of Lorraine | UdL · LORIA - Laboratoire Lorrain de Recherche en Informatique et Applications

The standard WSGG (Weighted Sum of Gray Gases) model is very fast and simple but can lead to relatively high discrepancies in the computation of the radiative heat transfer. A new model of high accuracy relying on the principle of the WSGG model is proposed. Contrary to the standard WSGG model, the pressure absorption coefficient depends here weakl...

The aim of the paper is to perform 3D radiative transfer calculations in combustion gas mixtures. For this purpose, a parallel modified Discrete Ordinates Method with a cell-vertex formulation on unstructured tetrahedral grid is proposed for solving the Radiative Transfer Equation (RTE) in realistic geometries. Several recent formulations of the we...

In optical tomography, the reconstructions have only been limited to the absorption μ a and scattering μ s coefficients of biological tissues due to theoretical and computational limitations. In this study, The authors propose an efficient method to reconstruct, in 3D geometries, the anisotropy factor g of the Henyey-Greenstein phase function as a...

Pseudo-Random Number Generators (PRNG) are omnipresent in computer science: they are embedded in all approaches of numerical simulation (for exhaustiveness), optimization (to discover new solutions), testing (to detect bugs) cryptography (to generate keys), and deep learning (for initialization, to allow generalizations). . . . PRNGs can be basical...

Purpose Near‐infrared optical imaging aims to reconstruct the absorption μa and scattering μs coefficients in order to detect tumors at early stage. However, the reconstructions have only been limited to μa and μs due to theoretical and computational limitations. The authors propose an efficient method of the reconstruction, in three‐dimensional ge...

In previous works, some of the authors have proposed a canonical form of Gray Codes (GCs) in N-cubes (hypercubes of dimension N). This form allowed them to draw an algorithm that theoretically provides exactly all the GCs for a given dimension N. In another work, we first have shown that any of these GC can be used to build the transition function...

In this paper, we suggest a different methodology to shorten the codeoptimization development time while getting a unified code with good performanceon different targeted devices. In the scope of this study, experiments are illustratedon a Discontinuous Galerkin code applied to Computational Fluid Dynamics. Testsare performed on CPUs, KNL Xeon-Phi...

Learn a simple strategy guideline to optimize applications runtime. The strategy is based on four steps and illustrated on a two-dimensional Discontinuous Galerkin solver for computational fluid dynamics on structured meshes. Starting from a CPU sequential code, we guide the audience through the different steps that allowed us to increase performan...

In this research, we suggest a different methodology to shorten the code optimization development time while getting a unified code with good performance on different targeted devices. In the scope of this study, experiments are illustrated on a Discontinuous Galerkin code applied to Computational Fluid Dynamics. Tests are performed on CPUs, KNL Xe...

Learn a simple strategy guideline to optimize applications runtime. The strategy is based on four steps and illustrated on a two-dimensional Discontinuous Galerkin solver for computational fluid dynamics on structured meshes. Starting from a CPU sequential code, we guide the audience through the different steps that allowed us to increase performan...

In previous works, the idea of walking into a \(\mathsf {N}\)-cube where a balanced Hamiltonian cycle have been removed has been proposed as the basis of a chaotic PRNG whose chaotic behavior has been proven. However, the construction and selection of the most suited balanced Hamiltonian cycles implies practical and theoretical issues. We propose i...

Designing a pseudorandom number generator (PRNG) is a difficult and complex task. Many recent works have considered chaotic functions as the basis of built PRNGs: the quality of the output would indeed be an obvious consequence of some chaos properties. However, there is no direct reasoning that goes from chaotic functions to uniform distribution o...

Hardware accelerators are classic scientific coprocessors in HPC machines. However, the number of CPU cores on the mother board is increasing and constitutes a non negligible part of the total computing power of the machine. So, running an application both on an accelerator (like a GPU or a Xeon-Phi device) and on the CPU cores can provide the high...

An efficient solution to detect tumor-like inclusions embedded within a human liver tissue model is presented, using illumination by a short-pulsed laser beam. Light propagation was accurately solved using the time-dependent radiative transfer equation, with multithreaded parallel computing. A modified finite volume method based on unstructured gri...

An important issue in tissue optics and Optical Tomography is to have an efficient forward solver. In this work, a new numerical algorithm was developed for solving light propagation with the radiative transport equation within a three-dimensional absorbing and a highly forward-scattering medium such as a biological tissue subjected to an incident...

We examine the accuracy of a modified finite volume method compared to analytical and Monte Carlo solutions for solving the radiative transfer equation. The model is used for predicting light propagation within a two-dimensional absorbing and highly forward-scattering medium such as biological tissue subjected to a collimated light beam. Numerical...

This chapter proposes to draw upon several development methodologies to obtain efficient codes in classical scientific applications. Those methodologies are based on the feedback from several research works involving GPUs, either in a single machine or in a cluster of machines. Indeed, our past collaborations with industries have allowed us to poin...

We present a near linear algorithm for determining the linear separability of two sets of points in a two-dimensional space. That algorithm does not only detects the linear separability but also computes separation information. When the sets are linearly separable, the algorithm provides a description of a separation hyperplane. For non linearly se...

We study the impact of asynchronism on parallel iterative algorithms in the particular context of local clusters of workstations including GPUs. The application test is a classical PDE problem of advection-diffusion-reaction in 3D. We propose an asynchronous version of a previously developed PDE solver using GPUs for the inner computations. The alg...

The purpose of this work is to further study the relevance of accelerating the Monte-Carlo calculations for the gamma rays external radiotherapy through feed-forward neural networks. We have previously presented a parallel incremental algorithm that builds neural networks of reduced size, while providing high quality approximations of the dose depo...

The purpose of this work is to further study the relevance of accelerating the Monte Carlo calculations for the gamma rays external radiotherapy through feed-forward neural networks. We have previously presented a parallel incremental algorithm that builds neural networks of reduced size, while providing high quality approximations of the dose depo...

Load balancing is a well known problem, which has been extensively addressed in parallel algorithmic. However, there subsist some contexts in which the existing algorithms cannot be used. One of these contexts is the case of dynamic networks where the links between the different elements are intermittent. We propose in this paper an efficient algor...

We present an experimental comparison between a synchronous and an asynchronous version of a same PDE solver on a GPU-cluster. In the context of our experiments, the GPU-cluster can be heterogeneous (different CPUs and GPUs). The comparison is done both on performance and energetic aspects.

We present a new sparse linear solver for GPUs. It is designed to work with structured sparse matrices where all the non-zeros are on a few diagonals. Several iterative algorithms are implemented, both on CPU and GPU. The GPU code is designed to be fast yet simple to read and understand. It aims to be as accurate as possible, even on chips that do...

We present a new sparse linear solver for GPUs. It is designed to work with structured sparse matrices where all the non-zeros are on a few diagonals. Several iterative algorithms are implemented, both on CPU and GPU. The GPU code is designed to be fast yet simple to read and understand. It aims to be as accurate as possible, even on chips that do...

This paper brings a correction to the formulation of the basins of fixed-point states of fully asynchronous discrete-time discrete-state dynamic networks presented in our paper that appeared in the IEEE Transactions on Neural Networks, vol. 17, no. 2, pp. 397-408, March 2006. In our subsequent works on totally asynchronous systems, we have discover...

This paper presents an incremental learning algorithm for feed-forward neural networks used as approximators of real world data. This algorithm allows neural networks of limited size to be obtained, providing better performances. The algorithm is compared to two of the main incremental algorithms (Dunkin and cascade correlation) in the respective c...

Présentation orale de l'adaptation de plusieurs algorithmes de résolution de systèmes linéaires creux sur GPU et de leur comparaison en termes de performance et précisison.

Présentation orale faisant dans un premier temps la synthèse des réqultats relatifs aux algorithmes itératifs asynchrones sur les systèmes de calcul de type grappes de machines locales ou réparties géographiquement. La seconde partie est une argumentation pour étudier l'adaptation de ces algorithmes sur grappe de GPUs.

This paper presents a parallel and fault tolerant version of an incremental learning algorithm for feed-forward neural networks used as function approximators. It has been shown in previous works that our incremental algorithm builds networks of reduced size while providing high quality approximations for real data sets. However, for very large set...

In this paper we present a practical, efficient and robust algorithm for detecting the global convergence in any asynchronous iterative process. A proven theoretical version, together with a first practical version, was presented in [1]. However, the main drawback of that first practical version was to require the determination of the maximal commu...

Direct dose calculations are a crucial requirement for Treatment Planning Systems. Some methods, such as Monte Carlo, explicitly model particle transport, others depend upon tabulated data or analytic formulae. However, their computation time is too lengthy for clinical use, or accuracy is insufficient, especially for recent techniques such as Inte...

Focusing on grid computing and asynchronism, Parallel Iterative Algorithms explores the theoretical and practical aspects of parallel numerical algorithms. Each chapter contains a theoretical discussion of the topic, an algorithmic section that fully details implementation examples and specific algorithms, and an evaluation of the advantages and dr...

An efficient and accurate algorithm for radiation dose evaluation is presented in this paper. Such computations are useful in the radiotherapic treatment planning of tumors. The originality of our approach is to use a neural network which has been trained with several homogeneous environments to deduce the doses in any kind of environment (possibly...

This paper gives a formulation of the basins of fixed point states of fully asynchronous discrete-time discrete-state dynamic networks. That formulation provides two advantages. The first one is to point out the different behaviors between synchronous and asynchronous modes and the second one is to allow us to easily deduce an algorithm which deter...

AIAC algorithms (Asynchronous Iterations Asynchronous Communications) are a particular class of parallel iterative algorithms. Their asynchronous nature makes them more efficient than their synchronous counterparts in numerous cases as has already been shown in previous works. The first goal of this article is to compare several parallel programmin...

Finite discrete-state discrete-time dynamic networks are used to model numerous complex dynamical systems. Nevertheless, a major problem with these networks concerns their use in the asynchronous mode. Effectively, as the dynamic is changed, the stabilization of the state of the network is not ensured any more in the asynchronous mode although it i...

The subject of this paper is to show the high efficiency of asynchronism for parallel iterative algorithms in the context of grid computing, that is to say, with machines scattered on a broad geographical scale. The question is: does asynchronism help to reduce the communication penalty and the overall computation time of such a given algorithm? Th...

In a previous paper, we have shown the very high power of asynchronism for parallel iterative algorithms in a global context of grid computing. In this article, we study the interest of coupling load balancing with asynchronism in such algorithms. After proposing a noncentralized version of dynamic load balancing which is best suited to asynchronis...

The main goal of external beam radiotherapy is the treatment of tumours, while sparing, as much as possible, surrounding healthy tissues. In order to master and optimize the dose distribution within the patient, dosimetric planning has to be carried out. Thus, for determining the most accurate dose distribution during treatment planning, a compromi...

We introduce a theoretical algorithm and its practical version to perform a decentralized detection of the global convergence of parallel asynchronous iterative algorithms. We prove that, even if the algorithm is completely decentralized, the detection of global convergence is achieved on one processor under the classical conditions. The proposed a...

AIAC algorithms (Asynchronous Iterations Asynchronous Communications) are a particular class of parallel iterative algorithms. Their asynchronous nature makes them more efficient than their syn- chronous counterparts in numerous cases as has already been shown in previous works. The first goal of this article is to compare several parallel programm...

In a previous work, we have shown the very high power of asynchronism for parallel iterative algorithms in a global context of grid computing. In this article, we study the interest of coupling load balancing with asynchronism in these algorithms. We propose a non-centralized version of dynamic load balancing which is best suited to asynchronism. A...

In Ecole d'hiver GRID 2002, Aussois, France

This report presents the architecture and the algorithms used in (Distributed Interactive Engineering Toolbox), a hierarchical set of components to build Network Enabled Server applications in a Grid environment. This environment is built on top of different tools which are able to locate an appropriate server depending of the client's request, the...

We consider networks of a large number of neurons (or units, processors, ...), whose dynamics are fully asynchronous with overlapping updating. We suppose that the neurons take a finite number of states (discrete states), and that the updating scheme is discrete in time. We make no hypotheses on the activation function of the neurons; the networks...

The subject of this paper is to show the very high power of asynchronism for iterative algorithms in the context of global computing, that is to say, with machines scattered all around the world. The question is whether or not asynchronism helps to reduce the communication penalty and the overall computation time of a given parallel algorithm. The...

In this paper, we present the parallelization of a Matlab-like tool called Scilab using high performance numerical libraries and dierent approaches based either on the duplication of Scilab processes or using computational servers. This tool allows to perform high level operations on distributed matrices in a metacomputing environment. We also pres...

In this paper, we present the developments realized in the OURAGAN project around the parallelization of a MATLAB-like tool called SCILAB. These developments use high performance numerical libraries and different approaches based either on the duplication of SCILAB processes or on computational servers. This tool, SCILAB, allows users to perform hi...

In this paper, we present the developments realized in the Ouragan project around the parallelization of a Matlab-like tool called Scilab. These developments use high-performance numerical libraries and different approaches based either on the duplication of Scilab processes or on computational servers. This tool, Scilab//, allows users to perform...

We present in this paper, a method for vectorization, matching and simplification of image contours in aerial stereovision. The goal is to compute a 3D reconstruction of the scene. The advantage of our method is that it only requires the extracted bitmap contours in one image of the couple. This is quite interesting since bitmap contours extraction...

Scilab is an application for numerical algebraic computations developed in the Meta-2 INRIA project. This program contains a toolbox allowing the realization of complex simulations with high transparency for the user. Nevertheless, computation times and data sizes may be quite large. Thus it seems interesting to provide an access to the power of pa...

This paper introduces and compares three parallel algorithms to compute general geometric image transformations on MIMD machines. We propose three variants of a parallel general scheme. We focus on the load balancing and the data redistributions. Experimental results are reported and compared. The implementation has been done using PPCM library all...

L'étude réalisée dans cette thèse met en relation deux domaines scientifiques, a priori distincts, que sont la géologie et l'informatique. En effet, le contexte de ce travail est de concevoir une chaîne complète de traitements parallèles sur les images satellites allant de la reconstruction tridimensionnelle à la visualisation des terrains ainsi re...

We propose in this paper, a parallel implementation of a ground visualization algorithm. Our input data consist in a Digital Elevation Model (DEM) covering a rectangular region, together with a raster image of the same area (the texture). The goal of the algorithm is to compute in parallel, images of the DEM from any point of view while mapping the...

We propose in this paper, a parallel implementation of a ground visualization algorithm. Our input data consist in a Digital Elevation Model (DEM) covering a rectangular region, together with a raster image of the same area (the texture). The goal of the algorithm is to compute in parallel, images of the DEM from any point of view while mapping the...

We consider networks of a large number of neurons (or units, processors, ...), whose dynamics are fully asynchronous with overlapping updating. We suppose that the neurons take a finite number of states (discrete states) and that the updating scheme is discrete in time. We make no hypotheses on the activation function of the neurons, the networks m...

Multiresolution approaches provide a powerful tool for image process- ing. In this report,we propose an overview of dieren t types of multiresolu- tion. We show this methodology is composed of two main complementary steps, the bottom-up and the top-down strategies. We also present some important existing applications like ltering and features extra...

The subject of this paper is to show the high eciency of asynchronism for parallel iterative algorithms in the context of meta-clusters, that is to say, with machines scattered on a broad geographical scale. The question is: does asynchronism help to reduce the communication penalty and the overall computation time of such a given algorithm ? The a...