Fast And Accurate Finite Element Multigrid Solvers For Pde Simulations On Gpu Clusters

This dissertation demonstrates that graphics processors (GPUs) as representatives of emerging many-core architectures are very well-suited for the fast and accurate solution of large, sparse linear systems of equations, using parallel multigrid methods on heterogeneous compute clusters. Such systems arise for instance in the discretisation of (elliptic) partial differential equations with finite elements. Fine-granular parallelisation techniques and methods to ensure accuracy are developed that enable at least one order of magnitude speedup over highly-tuned conventional CPU implementations, without sacrificing neither accuracy nor functionality.

The hybrid/heterogeneous nature of future microprocessors and large high-performance computing systems will result in a reliance on two major types of components: multicore/manycore central processing units and special purpose hardware/massively parallel accelerators. While these technologies have numerous benefits, they also pose substantial perfo

The two volume set LNCS 7133 and LNCS 7134 constitutes the thoroughly refereed post-conference proceedings of the 10th International Conference on Applied Parallel and Scientific Computing, PARA 2010, held in Reykjavík, Iceland, in June 2010. These volumes contain three keynote lectures, 29 revised papers and 45 minisymposia presentations arranged on the following topics: cloud computing, HPC algorithms, HPC programming tools, HPC in meteorology, parallel numerical algorithms, parallel computing in physics, scientific computing tools, HPC software engineering, simulations of atomic scale systems, tools and environments for accelerator based computational biomedicine, GPU computing, high performance computing interval methods, real-time access and processing of large data sets, linear algebra algorithms and software for multicore and hybrid architectures in honor of Fred Gustavson on his 75th birthday, memory and multicore issues in scientific computing - theory and praxis, multicore algorithms and implementations for application problems, fast PDE solvers and a posteriori error estimates, and scalable tools for high performance computing.

This state-of-the-art survey features topics related to the impact of multicore, manycore, and coprocessor technologies in science and for large-scale applications in an interdisciplinary environment. The papers cover issues of current research in mathematical modeling, design of parallel algorithms, aspects of microprocessor architecture, parallel programming languages, hardware-aware computing, heterogeneous platforms, manycore technologies, performance tuning, and requirements for large-scale applications. The contributions presented in this volume offer a survey on the state of the art, the concepts and perspectives for future developments. They are an outcome of an inspiring conference conceived and organized by the editors at the Karlsruhe Institute Technology (KIT) in September 2011. The twelve revised full papers presented together with two contributed papers focus on combination of new aspects of microprocessor technologies, parallel applications, numerical simulation, and software development; thus they clearly show the potential of emerging technologies in the area of multicore and manycore processors that are paving the way towards personal supercomputing and very likely towards exascale computing.

This book constitutes the thoroughly refereed post-conference proceedings of the International Dagstuhl-Seminar on Efficient Algorithms for Global Optimization Methods in Computer Vision, held in Dagstuhl Castle, Germany, in November 2011. The 8 revised full papers presented were carefully reviewed and selected by 12 lectures given at the seminar. The seminar focused on the entire algorithmic development pipeline for global optimization problems in computer vision: modelling, mathematical analysis, numerical solvers and parallelization. In particular, the goal of the seminar was to bring together researchers from all four fields to analyze and discuss the connections between the different stages of the algorithmic design pipeline.

This book gathers outstanding papers presented at the European Conference on Numerical Mathematics and Advanced Applications (ENUMATH 2019). The conference was organized by Delft University of Technology and was held in Egmond aan Zee, the Netherlands, from September 30 to October 4, 2019. Leading experts in the field presented the latest results and ideas regarding the design, implementation and analysis of numerical algorithms, as well as their applications to relevant societal problems. ENUMATH is a series of conferences held every two years to provide a forum for discussing basic aspects and new trends in numerical mathematics and scientific and industrial applications, all examined at the highest level of international expertise. The first ENUMATH was held in Paris in 1995, with successive installments at various sites across Europe, including Heidelberg (1997), Jyvaskyla (1999), lschia Porto (2001), Prague (2003), Santiago de Compostela (2005), Graz (2007), Uppsala (2009), Leicester (2011), Lausanne (2013), Ankara (2015) and Bergen (2017).

With recent changes in multicore and general-purpose computing on graphics processing units, the way parallel computers are used and programmed has drastically changed. It is important to provide a comprehensive study on how to use such machines written by specialists of the domain. The book provides recent research results in high-performance computing on complex environments, information on how to efficiently exploit heterogeneous and hierarchical architectures and distributed systems, detailed studies on the impact of applying heterogeneous computing practices to real problems, and applications varying from remote sensing to tomography. The content spans topics such as Numerical Analysis for Heterogeneous and Multicore Systems; Optimization of Communication for High Performance Heterogeneous and Hierarchical Platforms; Efficient Exploitation of Heterogeneous Architectures, Hybrid CPU+GPU, and Distributed Systems; Energy Awareness in High-Performance Computing; and Applications of Heterogeneous High-Performance Computing. • Covers cutting-edge research in HPC on complex environments, following an international collaboration of members of the ComplexHPC • Explains how to efficiently exploit heterogeneous and hierarchical architectures and distributed systems • Twenty-three chapters and over 100 illustrations cover domains such as numerical analysis, communication and storage, applications, GPUs and accelerators, and energy efficiency

This book constitutes the thoroughly refereed post-conference proceedings of the 9th International Conference on Large-Scale Scientific Computations, LSSC 2013, held in Sozopol, Bulgaria, in June 2013. The 74 revised full papers presented together with 5 plenary and invited papers were carefully reviewed and selected from numerous submissions. The papers are organized in topical sections on numerical modeling of fluids and structures; control and uncertain systems; Monte Carlo methods: theory, applications and distributed computing; theoretical and algorithmic advances in transport problems; applications of metaheuristics to large-scale problems; modeling and numerical simulation of processes in highly heterogeneous media; large-scale models: numerical methods, parallel computations and applications; numerical solvers on many-core systems; cloud and grid computing for resource-intensive scientific applications.