The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
In this paper, a novel and efficient implementation of the Marching Cubes (MC) algorithm is presented for the reconstruction of anatomical structures from real three-dimensional medical data. The proposed approach is based on a generic rule, able to triangulate all 15 standard cube configurations used in the classical MC algorithm as well as additional cases presented in the literature. The proposed...
We survey some of the recent research in developing multilevel algebraic solvers for elliptic problems. A key concept is the design of a hierarchy of coarse spaces and related interpolation operators which together satisfy certain approximation and stability properties to ensure the rapid convergence of the resulting multigrid algorithms. We will discuss smoothed agglomeration methods, harmonic extension...
In the paper we present the Chimera overlapping mesh technique applied to the solution of compressible flow problems. This technique is used to facilitate the grid generation in complex geometries allowing at the same time for natural parallelisation of the problem. The presented algorithm is particularly suitable for cases with large number of meshes overlapping in almost arbitrary manner (including...
This paper deals with parallel computation in electromagnetics. The boundary integral method has been developed to solve scattering by a perfect electric conducting or perfect dielectric bodies. Only parallel computation enables to modelize large devices. We present the implementation of the code on a distributed memory parallel machine. We focus in particular on the parallelization of the BiCGStab...
This paper discusses the approach that is being taken for parallelising the WAQUA/Kalman software of the Dutch Institute for Coastal and Marine Management. This software is used for numerical simulation of flow and transport phenomena in coastal waters and incorporates a Kalman filtering procedure to assimilate observational data into the simulation. In particular the Kalman filtering part of the...
This paper details the development of a multi-block parallel structured CFD solver to predict the fluid flow in Fuel Cells, thereby facilitating its use as a design tool. A multi-block parallel code is quicker, has greater memory efficiency and has better geometry conformity than a single block code. The multi-block approach allows geometry to be altered easily with only the changed blocks requiring...
The parallel implementation of GCR is addressed, with particular focus on communication costs associated with orthogonalization processes. This consideration brings up questions concerning the use of Householder reflections with GCR. To precondition the GCR method a block Gauss-Jacobi method is used. Approximate solvers are used to obtain a solution of the diagonal blocks. Experiments on a cluster...
In this paper parallel modelling of linear inhomogeneous 1D distributed parameter systems is considered. Two modelling methods in relation to parallel simulation are discussed: a two-point series expansion method and an eigenfunction expansion method. Both methods are validated with two different source terms for which the analytical solutions are known. A comparison of both approaches with respect...
From improved crash simulation to acoustic optimisation to innovative company-saving designs, distributed computing and meta-applications are enabling European industry to compete more effectively in many areas. Three such meta-applications are PROMENVIR, Optimus and TOOLSHED, all developed in recent ESPRIT projects. The PROMENVIR product provides users with all the functionality needed to perform...
This work presents a parallel version of a complex numerical algorithm to solve a lubrication problem studied in Tribology. The execution of the sequential algorithm on a workstation requires a huge amount of CPU time and memory resources. So, in order to reduce the computational cost, we have applied parallelization techniques to the most costly parts of the original source code. Some blocks of the...
This paper describes restructuring techniques for out-of-core programs (i.e., those that deal with very large quantities of data) based on exploiting locality using a combination of loop and data transformations. Writing efficient out-of-core program is an arduous task. As a result, compiler optimizations directed at improving I/O performance are becoming increasingly important. We describe how a...
The PaLaDiN (PArallel LArge Data set In Network of workstations) project is concerned with parallel out-of-core application running on cluster of workstations or PCs. In such architectures, each node has a virtual memory manager and a first idea is to use this feature to run “parallel out-of-core” application as a parallel in-core one. The out-of-core part of the problem, i.e. the schedule of data...
Due to the convergence of the fast microprocessors with low latency and high bandwidth communication networks, clusters of workstations are being used for high-performance computing. In this paper we present the design and implementation of a runtime system to support irregular applications on cluster of workstations, called “Collective I/O Clustering”. The system provides a friendly programming model...
Research in the parallel and scientific computing area has begun to focus on the development of Java-based programming environments. This paper describes the design of an original object-oriented database-style repository interface for high performance storage and retrieval of scientific data. The design is based on the standard interface to object databases that has been defined by the Object Database...
This paper describes CAMELot, a parallel cellular automata (CA) simulation environment developed as part of the Esprit project COLOMBO. The objective of this project is to apply parallel computing to the simulation of the in situ bioremediation of contaminated soils through the use of CA models. CAMELot provides a programming environment for the development, interactive control and visualisation of...
A new approach of parallel implementation of FDTD method on a novel software DSM (Distributed Shared Memory) system is proposed and implemented in this paper. Two unique advantages of our approach are:(1) it is easy to parallelize sequential FDTD codes, (2) it is possible to run programs which requires a memory space larger than the main memory of a single node.
Recent performance enhancements, upgrade applications, and rapid advances in computers and networking technologies have all enabled information processing devices to share computer resources. We are interested in image processing that requires a large amount of calculations. If individual pixels forming an image can be calculated independently, the processing speed can be increased by parallel processing,...
Some of the current microprocessors provide a prefetch instruction, but either the instruction is treated as a NOP, (e.g. Digital Alpha EV4/5), or only a small number of outstanding prefetches is permitted (e.g. MIPS R10K). This paper discusses the design and implementation of the hardware support required to fully support the prefetch instruction for the Digital Alpha architecture. The prefetch support...
This paper studies the one-to-all broadcast in a circuit-switched 2D torus of any size with α-port capability. This is a generalization of the one-port and all-port models. Existing results, as compared to ours, can only solve very restricted sizes of tori, and use more numbers of steps.
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.