Boguslaw Butrylo

A survey of parallel solvers for the finite element method in computational electromagnetics

This paper presents new trends in parallel methods used to solve finite element matrix systems: standard iterative and direct solving methods first, and then domain decomposition methods. For example, the current status and properties of two prevailing programming environments (PVM and MPI) are finally given and compared when implemented together with a finite element time domain formulation.

Parallel SSOR preconditioning implemented on dynamic SMP clusters with communication on the fly

The paper presents a comparative analysis of parallel implementation of the preconditioned conjugate gradient method with the symmetric-successive over-relaxation preconditioner. Two parallel implementations of the matrix solver are compared. The first one is a message-passing version executed on a cluster of workstations. The other one is an efficient version simulated on a novel architecture of dynamically reconfigurable shared memory clusters with a new paradigm of inter-processor communication called communication on the fly. The presented example shows high suitability of the proposed architecture for fine grain numerical computations. It can be very useful in the simulation of physical phenomena described as numerical problems suitable for fine grain parallel execution.

Parallel broadband finite element time domain algorithm implemented to dispersive electromagnetic problem

The numerical analysis of some broadband electromagnetic fields and frequency-dependent materials using a time domain method is the main subject of this paper. The spatial and time-dependent distribution of the electromagnetic field is approximated by the finite element method. The parallel form of the algorithm valid for some linear materials, and the formulation of the FE code for a dispersive electromagnetic problem are presented and compared. The complex forms of these algorithms have an effect on the memory and computational costs of the distributed formulation. The properties of the algorithm are estimated using high performance cluster of workstations.

SSOR Preconditioned Conjugate Gradient Algorithm in Dynamic SMP Clusters with Communication on the Fly

The paper concerns efficient parallel computation methods for solving matrix linear equations using the preconditioned conjugate gradient method. Efficient parallelization methods of the preconditioner module have been proposed based on execution in a special architecture of dynamically reconfigurable shared memory clusters. The assumed system is provided with a novel efficient inter-processor communication technique called communication on the fly, which is very suitable for fine grain parallel computations. Two-step parallel data decomposition is implemented in the elaborated preconditioner, which leads to the decrease of computation grain and overlapping of communication with computations. Comparative experimental results are presented of program execution efficiency in a standard distributed multicomputer cluster based on message passing and in the architecture of dynamic SMP clusters with the communication on the fly

Limits of the Distributed Finite Element Time Domain Algorithm in Multi-Computer Environment

This paper deals with a concurrent version of the finite element time domain method. The algorithm works in SPMD mode and is executed in the MPI distributed environment. Its properties are validated with some different multi-computer, loosely coupled platforms. The real value and approximated asymptotic values of the basic metrics of speedup are presented and discussed.

The Parallel FDTD Algorithm and Its Application to the Analysis of Complex Structures

The properties of the approximated methods applied in the analysis of electromagnetic phenomena are presented. The detailed description of restrictions and advantages of the FDTD method is given. The principles of parallel implementation and the properties of the distributed FDTD algorithm are discussed. Analysis of the parallel processing performance is carried out. The presented parallel FDTD algorithm calculates the propagation of electromagnetic wave in the semi-infinite space.

Simulation of the electromagnetic vector field in the high-frequency components with dispersive materials

This paper deals with parallel analysis of an electromagnetic problem for frequency-dependent materials. The solution of the linear wave equation is achieved using the finite element time domain method. A multi-pole Debye model approximates the memory of the dispersive material. The time-domain convolution is explicitly calculated using the integration by parts method. The linear recursive technique is implemented to estimate the convolution integral in each time step. The comparative study of two parallel implementations of the algorithm is presented. The discussed parallel versions are based on the domain decomposition paradigm and task decomposition scheme. The task decomposition is developed using some heuristic scheme of coupling of parallel tasks. The presented algorithms are executed over a distributed, homogeneous testbed. Some aspects connected with simulation of a broadband electromagnetic field in a dispersive and isotropic material are discussed.

Granular SSOR Preconditioning Placed on Dynamic SMP Clusters with Communication on the Fly

The paper presents a comparative analysis of parallel implementation of the preconditioned conjugate gradient method with symmetric successive over relaxation preconditioner. Two parallel implementations are compared. The first one is a message passing version that turned out to be inefficient when executed on a typical cluster of workstations. The other one is an efficient version simulated on a novel architecture of dynamically reconfigurable shared memory clusters with a new paradigm of inter-processor communication called communication on the fly. The presented example shows high suitability of the proposed architecture for fine grain numerical computations what can be very useful in simulation of physical phenomena described as numerical problems for fine grain parallel execution.

Distributed Formulation of Artificial Reconstruction Technique with Reordering of Critical Data Sets

This paper discusses iterative algorithms for reconstruction of 2D images. The formulation of the distributed algorithm and its properties are described. The details of the distributed implementation of the ART (algebraic reconstruction technique) are presented. Parallel implementation in a parallel system based on dynamically configurable clusters is proposed

Temperature distribution in electrolytic cell and its impact on the electroplating process

A method of analysing the distribution of temperature fields in electrolytic cell is described. The finite volume method, a kind of approximated numerical method, is used to solve a complex and dual-field problem, where the mass transport of the circulating electrolyte is coupled with effects of transport of heat energy. A specific correct distribution of the temperature around the cathode plays a significant role in the electroplating process. This problem is discussed in the paper. Some useful remarks concerning of the heat sources and thermal field in the electrolyte are presented. As an example of the use of these methods, a test model is analysed. The comparative results are obtained for the reduced two-dimensional models.