Abstract: "This dissertation presents new techniques for solving large sparse unsymmetric linear systems on high performance computers, using Gaussian elimination with partial pivoting. The efficiencies of the new algorithms are demonstrated for matrices from various fields and for a variety of high performance machines. In the first part we discuss optimizations of a sequential algorithm to exploit the memory hierarchies that exist in most RISC-based superscalar computers. We begin with the left-looking supernode-column algorithm by Eisenstat, Gilbert and Liu, which includes Eisenstat and Liu's symmetric structural reduction for fast symbolic factorization. Our key contribution is to develop both numeric and symbolic schemes to perform supernode-panel updates to achieve better data reuse in cache and floating-point registers. A further refinement, a two-dimensional matrix partitioning scheme, enhances performance for large matrices or machines with small caches. We conduct extensive performance evaluations on several recent superscalar architectures, such as the IBM RS/6000-590, MIPS R8000 and DEC Alpha 21164, and show that our algorithm is much faster than its predecessors. The advantage is particularly evident for large problems. In addition, we develop a detailed model to systematically choose a set of blocking parameters in the algorithm. The second part focuses on the design, implementation and performance analysis of a shared memory parallel algorithm based on our new serial algorithm. We parallelize the computation along the column dimension of the matrix, assigning one block of columns (a panel) to a processor. The parallel algorithm retains the serial algorithm's ability to reuse cached data. We develop a dynamic scheduling mechanism to schedule tasks onto available processors. One merit of this approach is the ability to balance work load automatically. The algorithm attempts to schedule independent tasks to different processors. When this is not possible in the later stage of factorization, a pipeline approach is used to coordinate dependent computations. We demonstrate that the new parallel algorithm is very efficient on shared memory machines with modest numbers of processors, such as the SGI Power Challenge, DEC AlphaServer 8400, and Cray C90/J90. We also develop performance models to study available concurrency and identify performance bottlenecks."

This book constitutes the thoroughly refereed post-conference proceedings of the 8th International Conference on High Performance Computing for Computational Science, VECPAR 2008, held in Toulouse, France, in June 2008. The 51 revised full papers presented together with the abstract of a surveying and look-ahead talk were carefully reviewed and selected from 73 submissions. The papers are organized in topical sections on parallel and distributed computing, cluster and grid computing, problem solving environment and data centric, numerical methods, linear algebra, computing in geosciences and biosciences, imaging and graphics, computing for aerospace and engineering, and high-performance data management in grid environments.

This book constitutes the thoroughly refereed post-conference proceedings of the 11th International Conference on High Performance Computing for Computational Science, VECPAR 2014, held in Eugene, OR, USA, in June/July 2014. The 25 papers presented were carefully reviewed and selected of numerous submissions. The papers are organized in topical sections on algorithms for GPU and manycores, large-scale applications, numerical algorithms, direct/hybrid methods for solving sparse matrices, performance tuning. The volume also contains the papers presented at the 9th International Workshop on Automatic Performance Tuning.

Provides a rapid introduction to the world of vector and parallel processing for these linear algebra applications.

The Leibniz Supercomputing Centre (LRZ) and the Bavarian Competence Network for Technical and Scienti?c High Performance Computing (KONWIHR) publish in the present book results of numerical simulations facilitated by the High P- formance Computer System in Bavaria (HLRB II) within the last two years. The papers were presented at the Fourth Joint HLRB and KONWIHR Review and - sult Workshop in Garching on 8th and 9th December 2009, and were selected from all progress reports of projects that use the HLRB II. Similar to the workshop two years ago, the majority of the contributed papers belong to the area of computational ?uid dynamics (CFD), condensed matter physics, astrophysics, chemistry, computer sciences and high-energy physics. We note a considerable increase of the user c- munity in some areas: Compared to 2007, the number of papers increased from 6 to 12 in condensed matter physics and from 2 to 5 in high-energy physics. Bio s- ences contributed only one paper in 2007, but four papers in 2009. This indicates that the area of application of supercomputers is continuously growing and entering new ?elds of research. The year 2007 saw two major events of particular importance for the LRZ. First, after a substantial upgrade with dual-core processors the SGI Altix 4700 superc- puter reached a peak performance of more than 62 Tera?op/s. And second, the n- pro?t organization Gauss Centre for Supercomputing e. V. (GCS) was founded on April 13th.

This book constitutes the refereed proceedings of the 4th International Conference on Parallel Computation, ACPC'99, held in Salzburg, Austria in February 1999; the conference included special tracks on parallel numerics and on parallel computing in image processing, video processing, and multimedia. The volume presents 50 revised full papers selected from a total of 75 submissions. Also included are four invited papers and 15 posters. The papers are organized in topical sections on linear algebra, differential equations and interpolation, (Quasi-)Monte Carlo methods, numerical software, numerical applications, image segmentation and image understanding, motion estimation and block matching, video processing, wavelet techniques, satellite image processing, data structures, data partitioning, resource allocation and performance analysis, cluster computing, and simulation and applications.

This volume constitutes the papers of several workshops which were held in conjunction with the 38th International Conference on High Performance Computing, ISC High Performance 2023, held in Hamburg, Germany, during May 21–25, 2023. The 49 revised full papers presented in this book were carefully reviewed and selected from 70 submissions. ISC High Performance 2023 presents the following workshops: ​2nd International Workshop on Malleability Techniques Applications in High-Performance Computing (HPCMALL) 18th Workshop on Virtualization in High-Performance Cloud Computing (VHPC 23) HPC I/O in the Data Center (HPC IODC) Workshop on Converged Computing of Cloud, HPC, and Edge (WOCC’23) 7th International Workshop on In Situ Visualization (WOIV’23) Workshop on Monitoring and Operational Data Analytics (MODA23) 2nd Workshop on Communication, I/O, and Storage at Scale on Next-Generation Platforms: Scalable Infrastructures First International Workshop on RISC-V for HPC Second Combined Workshop on Interactive and Urgent Supercomputing (CWIUS) HPC on Heterogeneous Hardware (H3)

This book constitutes the thoroughly refereed post-proceedings of the 7th International Conference on High Performance Computing for Computational Science, VECPAR 2006, held in Rio de Janeiro, Brazil, in June 2006. The 44 revised full papers presented together with one invited paper and 12 revised workshop papers cover Grid computing, cluster computing, numerical methods, large-scale simulations in Physics, and computing in Biosciences.

Leading researchers have contributed state-of-the-art chapters to this overview of high-performance computing in biomedical research. The book includes over 30 pages of color illustrations. Some of the important topics featured in the book include the following:

This book constitutes the thoroughly refereed post-proceedings of the Second International Conference on Numerical Analysis and Its Applications, NAA 2000, held in Rousse, Bulgaria in June 2000.The 90 revised papers presented were carefully selected for inclusion in the book during the two rounds of inspection and reviewing. All current aspects of numerical analysis are addressed. Among the application fields covered are computational sciences and engineering, chemistry, physics, economics, simulation, etc.