Papers by Achim Basermann
電気学会研究会資料 Mag マグネティックス研究会, Jun 22, 2000
Parallel Processing for Scientific Computing, 1995
Parallel Sparse Matrix Computations in Iterative Solvers on Distributed Memory Machines* Achim Ba... more Parallel Sparse Matrix Computations in Iterative Solvers on Distributed Memory Machines* Achim Basermannt Abstract Sparse matrix computations play an important ...
Lecture Notes in Computer Science, 2005
ABSTRACT A parallel version of the Generalized Finite Element Method is applied to multiparticle ... more ABSTRACT A parallel version of the Generalized Finite Element Method is applied to multiparticle problems. The main advantage of the method is that only a regular hexahedral grid is needed; the particles do not have to be meshed and are represented by special basis functions approximating the field behavior near the particles. A general-purpose parallel Schur complement solver with incomplete LU preconditioning (A. Basermann) showed excellent performance for the varying problem size, number of processors and number of particles. In fact, the scaling of the computational time with respect to the number of processors was slightly superlinear due to cache effects. Future research plans include parallel implementation of the new Flexible Local Approximation MEthod (FLAME) that incorporates desirable local approximating functions (e.g. dipole harmonics near particles) into the difference scheme.
International Journal of Modern Physics C, 1993
ABSTRACT
Lecture Notes in Computer Science, 1999
One important mathematical problem in simulation of large electrical circuits is the solution of ... more One important mathematical problem in simulation of large electrical circuits is the solution of high-dimensional linear equation systems. The corresponding matrices are real, non-symmetric, very ill-conditioned, have an irregular sparsity pattern, and include a few dense rows and columns.
The DRAMA project is a European Commission (ESPRIT)-funded project which has been initiated to su... more The DRAMA project is a European Commission (ESPRIT)-funded project which has been initiated to support the take-up of large scale parallel simulation in industry by dealing with one of the main problems which restricts the use of message-passing simulation codes - the inability to perform dynamic load balancing. A concentration on message-passing parallelisation corresponds to the target of addressing large scale and thus highly scalable parallel applications. The central product of the project will be a library comprising various tools for dynamic repartitioning of unstructured finite element applications. The particular focus of the project is on the requirements of industrial Finite Element codes, with library evaluation and validation being performed using industrial software packages, but codes using Finite Volume formulations will also be able to make use of the project results. The core library functions will perform a parallel computation of a mesh re-allocation that will re...
Vorgestellt werden iterative Krylov-Unterraum-Verfahren zur Lösung großer, dünnbesetzter linearer... more Vorgestellt werden iterative Krylov-Unterraum-Verfahren zur Lösung großer, dünnbesetzter linearer Gleichungssysteme auf Parallelrechnern mit verteiltem Speicher. Zur Beschleunigung der Verfahren werden Vorkonditionierungsmethoden eingesetzt, die auf einem Schur-Komplement-Ansatz beruhen. Diese können als eine iterative Verbesserung einer Block-Jacobi-Vorkonditionierung auf Basis unvollständiger LU-Zerlegungen von Diagonalblöcken der Matrix formuliert werden. Als Beispiele dienen Gleichungssysteme aus Anwendungen zur Strömungssimulation des Deutschen Zentrums für Luft- und Raumfahrt. Das Verhalten der Methoden zur Vorkonditionierung wird anhand empirischer Tests mit diesen Gleichungssystemen bezüglich der Laufzeit und der numerischen Eigenschaften erörtert. Performance-Untersuchungen zeigen, dass sich die parallele Skalierbarkeit des Lösungsverfahrens bei einer Schur-Komplement-Vorkonditionierung deutlich gegenüber einer Block-Jacobi-Vorkonditionierung verbessert. Dieses Verfahren lä...
Developments in Computational Mechanics with High Performance Computing, 1999
Developments in Engineering Computational Technology, 2000
Die interaktive Analyse großer Simulationsdatensätze soll durch skalierbares Multi-Resolution-Str... more Die interaktive Analyse großer Simulationsdatensätze soll durch skalierbares Multi-Resolution-Streaming ermöglicht werden. Da sehr große Simulationsdatensätze, die z.B. bei TRACE-Läufen erzeugt werden, nicht in kurzer Zeit voll-ständig verarbeitet werden können, sollen interaktive Visualisierungsmethoden entwickelt werden, die • durch Multi-Resolution-Datenstrukturen einen schnellen Zugriff auf Teildaten erlauben, • Many-Core-Systeme für ein effizientes, paralleles Postprocessing verwenden und • moderne GPU-Architekturen sowohl zur Unterstützung des Postprocessings als auch für eine qualitativ hochwertige Darstellung der Ergebnisse mit interakti-ven Bildwiederholungsraten nutzen. Das zentrale Ziel besteht darin, alle Stufen der kompletten Postprocessing-Pipeline permanent mit ausreichend Daten zu versorgen, um immer detailliertere Ergebnisse für die interaktive Analyse zu erzeugen.
Mathematics in Industry, 2004
ABSTRACT For solving sparse linear systems from circuit simulation whose coefficient matrices inc... more ABSTRACT For solving sparse linear systems from circuit simulation whose coefficient matrices include a few dense rows and columns, a parallel Bi-CGSTAB algorithm with distributed Schur complement preconditioning is presented. The parallel efficiency of the solver is increased by transforming the equation system into a problem without dense rows and columns as well as by exploitation of parallel graph partitioning methods. The costs of local, incomplete LU decompositions are decreased by fill-in reducing reordering methods of the matrix and a threshold strategy for the factorization. The efficiency of the parallel solver is demonstrated with real circuit simulation problems on a PC cluster.
project aims to bring together the developments in parallel partitioning and parallel FE applicat... more project aims to bring together the developments in parallel partitioning and parallel FE applications to ensure that the potential of scalable computing can be achieved for fully-functional industrial simulation, which includes efficient adaptive meshing (and re-meshing) options. The parallel dynamic re-partitioning routines should also be to handle the full complexity and range of finite elements as used in industrial structural mechanics codes, as exemplified by the applications within the project.
Advances in Computational Mechanics with High Performance Computing, 1998
Numerical Linear Algebra with Applications, 2000
Journal of Parallel and Distributed Computing, 1997
ABSTRACT
Uploads
Papers by Achim Basermann