Parallel Numerics 96 Home Page

Parallel Numerics provides a forum for the presenatation and exchange of current work on the workpackage WP.5 "Parallel Numerics" of the international project PACT "Programming Environments, Algorithms, Applications, Compilers and Tools for Parallel Computation." Research teams from six countries are taking part in this project: University of Salzburg and RISC Institute Linz, (Austria), Hungarian Academy of Sciences and Technical University of Budapest (Hungary), University of Bologna and IRSIP-CNR Institute Naples (Italy), University of Czestochowa (Poland) with associated partner Politechnic Instititute Kiev (Ukraine), Slovak Academy of Science and Technical University of Kosice (Slovakia), and Jozef Stefan Institute (Slovenia). The fist two workshops were held in Smolenice, Slovakia (1994) and Sorrento, Italy (1995). The third workshop was organized by Jozef Stefan Institute. The workshop took place in the Hotel Spik in Gozd Martuljek, a small village near Kranjska Gora, the famous ski resort in the north-western part of Slovenia.

ISBN 86-80023-25-6

Program Committee

Roman Trobec, Jozef Stefan Institute, Slovenia (Chair)
Katalin Balla, Hungarian Academy of Sciences, Hungary
Polona Blaznik, Jozef Stefan Institute, Slovenia
István Deák, Technical University of Budapest, Hungary
Giuseppe De Pietro, IRSIP National Research Council, Italy
Carla Guerrini, University of Bologna, Italy
Gorazd Kandus, Jozef Stefan Institute, Slovenia
István Maros, Hungarian Academy of Sciences, Hungary
Borut Robic, Jozef Stefan Institute, Slovenia
Jurij Silc, Jozef Stefan Institute, Slovenia
Sabine Stifter, University of Linz, Austria
Marian Vajtersic, Slovak Academy of Science, Slovakia
Roman Wyrzykowski, Technical University of Czestochwa, Poland
Hans P. Zima, University of Vienna, Austria
Peter Zinterhof, University of Salzburg, Austria

Organizing Committee

Polona Blaznik, Jozef Stefan Institute, Slovenia (Chair)
Dusan Caf, Jozef Stefan Institute, Slovenia
Bostjan Slivnik, Jozef Stefan Institute, Slovenia
Anastazija Ugrin, Jozef Stefan Institute, Slovenia (Secretary)

Workshop Program

Session A (Chair: R.Trobec)

P.Zinterhof: On Optimal Liftings of Sequences
W.Ch.Schmid: An Algorithm to Determine the Quality Parameter of Binary Nets, and the New Shift-Method
S.Wegenkittl, K.Entacher: On the Relevance of Splitting Properties and the Compound Method in Parallel Applications of Pseudorandom Number Generators
H.Piech, A.Borowik, T.Chlopas, N.Gubarieni: Verification of Theorem Thruthfulness in Three-Valued Logics with Parallel System Applications

Session B (Chair: M.Vajtersic)

H.P.Zima: High-Level Programming Support for Scalable Parallel Systems (Invited talk)
R.Wyrzykowski, J.S.Kanevski, V.L.Lepecha: Mapping Sparse Matrix Computations into Processor Arrays
R.Reinhardt: SiliconGraphics Multiprocessor Architectures

Session C (Chair: P.Zinterhof)

P.-E.Bernard, B. Plateau, D.Trystram: Using Threads for Developing Parallel Applications: Molecular Dynamics as a Case Study (Invited talk)
A.Uhl: Adaptive Quadtree Based Wavelet Packet Image Block-Coding on MIMD Parallel Architectures
J.Hämmerle: Combining Sequential Speed-up Techniques and Parallel Computing for Fractal Image Compression

Session D (Chair: I.Deak)

M.Vajtersic: Examples of Two-Sided Jacobi Algorithms for SVD Computation on Hypercubes, Rings and Meshes
M.Becka: An Implementation of the SVD on the Intel Paragon
P.Blaznik: Low-Rank Modification of Two-Sided Decompositions
R.Wyrzykowski, H.Piech, T.Chlopas, V.L.Lepecha: On an Algorithm for LU Factorization of Matrices

Session E (Chair: P.Blaznik)

D.Janezic, F.Merzel: Molecular Dynamics Simulations of Macromolecules: Improvements on Computational Efficiency (Invited talk)
L.Halada, M.Lucka, E.Viktorinova: Solving Block Pentadiagonal Linear System on MIMD Computer Systems

Session F (Chair:K.Balla)

D.Brunner, G.Larcher, W.Ch.Schmid: PANUMIWAL: A System for the Parallel Numerical Integration of High-Dimensional Functions
P.Schmitzberger: Approximation and Interpolation of High Dimensional Functions by Generalized Walsh Polynomials
B.Slivnik, R.Trobec, B.Orel: Comparison of 3- and 5-stage Multi-Implicit Runge-Kutta Methods

Session G (Chair: B.Robic)

V.Valencic: Computational Electromagnetics - Parallel Numerical Methods to Determine the Current Density Distributions inside Human Tissues (Invited talk)
P.Purcz: The Speed Up of a Parallel Algorithm for a Cauchy Problem and for Various Known Numerical Methods
K. Balla: On some issues of parallelization of boundary value problems for ODE

Session H (Chair: R.Wyrzykowski)

E.Hausenblas: A Parallelizable Algorithm for Solving Partial Differential Equations of Neumann Type Based on the Euler Scheme
I.Deak: Asynchronous Parallel Models for Optimization

Program Committee

Organizing Committee

Workshop Program

Sponsors

Gozd Martuljek, September 13, 1996