High Performance Computing Techniques Research Articles

Scalable parallel preconditioners for an open source cardiac electrophysiology simulation package

The numerical solution of the bidomain equations is a fundamental tool in the field of computational cardiac electrophysiology. The multi-scale nature of the processes involved and the increasing complexity of the tissue and ionic models of interest make the use of High Performance Computing techniques mandatory. The solution of the linear system arising from the Finite Element discretisation of the bidomain equations is the main bottleneck in terms of performance and parallel scalability. In this work we present and evaluate parallel implementations of two problemspecific block preconditioners, namely BD and LDU, in the framework of the open source cardiac electrophysiology simulation package Chaste. Both rely upon providing a reliable and practical method for calculating the action of the inverse of certain matrix subblocks. Our results show that: a) total execution time can be reduced by using different approximation methods for different matrix subblocks, b) there exists a good correlation between the coefficient Δt=h2 and the cases where BD and LDU outperform generic Algebraic Multigrid (AMG) preconditioning, and c) when applied to simulations with current state-of-the-art cardiac geometries, BD and LDU are over 3 times faster than AMG preconditioning and show similar scaling.

Acceleration of FDTD mode solver by high-performance computing techniques

A two-dimensional (2D) compact finite-difference time-domain (FDTD) mode solver is developed based on wave equation formalism in combination with the matrix pencil method (MPM). The method is validated for calculation of both real guided and complex leaky modes of typical optical waveguides against the bench-mark finite-difference (FD) eigen mode solver. By taking advantage of the inherent parallel nature of the FDTD algorithm, the mode solver is implemented on graphics processing units (GPUs) using the compute unified device architecture (CUDA). It is demonstrated that the high-performance computing technique leads to significant acceleration of the FDTD mode solver with more than 30 times improvement in computational efficiency in comparison with the conventional FDTD mode solver running on CPU of a standard desktop computer. The computational efficiency of the accelerated FDTD method is in the same order of magnitude of the standard finite-difference eigen mode solver and yet require much less memory (e.g., less than 10%). Therefore, the new method may serve as an efficient, accurate and robust tool for mode calculation of optical waveguides even when the conventional eigen value mode solvers are no longer applicable due to memory limitation.

Development of an Unsteady Aerodynamic Simulator Using Large-Eddy Simulation Based on High-Performance Computing Technique

Development of an Unsteady Aerodynamic Simulator Using Large-Eddy Simulation Based on High-Performance Computing Technique

DOVIS: an implementation for high-throughput virtual screening using AutoDock

BackgroundMolecular-docking-based virtual screening is an important tool in drug discovery that is used to significantly reduce the number of possible chemical compounds to be investigated. In addition to the selection of a sound docking strategy with appropriate scoring functions, another technical challenge is to in silico screen millions of compounds in a reasonable time. To meet this challenge, it is necessary to use high performance computing (HPC) platforms and techniques. However, the development of an integrated HPC system that makes efficient use of its elements is not trivial.ResultsWe have developed an application termed DOVIS that uses AutoDock (version 3) as the docking engine and runs in parallel on a Linux cluster. DOVIS can efficiently dock large numbers (millions) of small molecules (ligands) to a receptor, screening 500 to 1,000 compounds per processor per day. Furthermore, in DOVIS, the docking session is fully integrated and automated in that the inputs are specified via a graphical user interface, the calculations are fully integrated with a Linux cluster queuing system for parallel processing, and the results can be visualized and queried.ConclusionDOVIS removes most of the complexities and organizational problems associated with large-scale high-throughput virtual screening, and provides a convenient and efficient solution for AutoDock users to use this software in a Linux cluster platform.

High Performance Computing Techniques for the FEM Simulation in Structural Mechanics

AbstractFor the solution of large scale simulations in structural mechanics iterative solving methods are mandatory. The efficiency of such methods can crucially depend on different factors: choice of material parameters, quality of the underlying computational mesh and number of processors in a parallel computing system. We distinguish between three aspects of ‘efficiency’: processor efficiency (degree to which the solving algorithm is able to exploit the processor's computational power), parallel efficiency (ratio between computation and communication times) and numerical efficiency (convergence behaviour). With the new FEM software package Feast we pursue the aim to develop a solver mechanism which at the same time gains high efficiencies in all three aspects, while trying to minimise the mentioned dependencies. (© 2006 WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim)

Effective Permeability Calculation Using Boundary Element Method in Naturally Fractured Reservoirs

A model is presented to calculate the effective permeability tensor in naturally fractured reservoirs using Boundary Element Methods (BEM). Arbitrary fractures of different scales based on their length are considered. Interface boundary condition is used to model the short fractures as an enhancement of matrix permeability. Long fractures, on the other hand are treated as source/sink in the corresponding blocks. Periodic boundary condition is applied to the grid-block boundaries to calculate the elements of effective permeability tensor. Darcy's law and Navier-stoke's equation are applied to fluid flow in rock matrix and fractures, respectively. An important feature of this approach is that the fluid flow in matrix-fracture interface is coupled by Poisson's equation and fluid flow in the rest of the matrix is formulated by Laplace's equation. This paper also presents an innovative approach to optimization and parallelization of the model by High Performance Computing (HPC) techniques. The model has been validated against analytical results and applied to a typical case where arbitrary fractures of different sizes are assumed within the grid blocks. The effective block permeability tensors can be implemented into a reservoir simulator to calculate fluid flow through the naturally fractured reservoirs.

High-performance electron tomography of complex biological specimens

We have evaluated reconstruction methods using smooth basis functions in the electron tomography of complex biological specimens. In particular, we have investigated series expansion methods, with special emphasis on parallel computation. Among the methods investigated, the component averaging techniques have proven to be most efficient and have generally shown fast convergence rates. The use of smooth basis functions provides the reconstruction algorithms with an implicit regularization mechanism, very appropriate for noisy conditions. Furthermore, we have applied high-performance computing (HPC) techniques to address the computational requirements demanded by the reconstruction of large volumes. One of the standard techniques in parallel computing, domain decomposition, has yielded an effective computational algorithm which hides the latencies due to interprocessor communication. We present comparisons with weighted back-projection (WBP), one of the standard reconstruction methods in the areas of computational demand and reconstruction quality under noisy conditions. These techniques yield better results, according to objective measures of quality, than the weighted backprojection techniques after a very few iterations. As a consequence, the combination of efficient iterative algorithms and HPC techniques has proven to be well suited to the reconstruction of large biological specimens in electron tomography, yielding solutions in reasonable computation times.

Characterizing land surface anisotropy from AVHRR data at a global scale using high performance computing

We used the multi-temporal ten-day composite data from the Advanced Very High Resolution Radiometer (AVHRR) for the years 1983 to 1986 to retrieve the Bidirectional Ree ectance Distribution Function (BRDF) using high performance computing techniques. Three di Verent models are used: a simple linear model, a semi-empirical iterative model and a temporal model. The objectives of this study were to compare the performance of di Verent BRDF models at a global scale, assess the computational requirements and optimize the algorithm implementation using high performance computational techniques, and to determine ifthere is any coherent spatial structure in the coe Ycients ofdiVerent BRDF models corresponding to di Verent land cover types. The standard error between model computed ree ectances and the input data was used to quantify the performance ofthe models.Even though the iterative model is computationally more expensive (158 minutes) than either the simple linear model (15 minutes) or the temporal model (16 minutes), the results from all the three models were very similar when the BRDF was estimated at discrete time periods. If the BRDF models were applied without dividing the input data into discrete time intervals, then the temporal model gave better results than the other two. All the models were run on an IBM SP2 parallel machine with 16 CPUs. Most of the mountain- ous and snow covered areas in high latitudes had null values since the cloud screening algorithm used in the Pathe nder processing performed poorly in distin- guishing between snow and clouds. The BRDF coe Ycients of the iterative model and the Fourier coe Ycients of the temporal model showed a strong spatial structure corresponding to known variations in land cover.

Characterizing land surface anisotropy from AVHRR data at a global scale using high performance computing

We used the multi-temporal ten-day composite data from the Advanced Very High Resolution Radiometer (AVHRR) for the years 1983 to 1986 to retrieve the Bidirectional Reflectance Distribution Function (BRDF) using high performance computing techniques. Three different models are used: a simple linear model, a semi-empirical iterative model and a temporal model. The objectives of this study were to compare the performance of different BRDF models at a global scale, assess the computational requirements and optimize the algorithm implementation using high performance computational techniques, and to determine if there is any coherent spatial structure in the coefficients of different BRDF models corresponding to different land cover types. The standard error between model computed reflectances and the input data was used to quantify the performance of the models. Even though the iterative model is computationally more expensive (158 minutes) than either the simple linear model (15 minutes) or the temporal model (16 minutes), the results from all the three models were very similar when the BRDF was estimated at discrete time periods. If the BRDF models were applied without dividing the input data into discrete time intervals, then the temporal model gave better results than the other two. All the models were run on an IBM SP2 parallel machine with 16 CPUs. Most of the mountainous and snow covered areas in high latitudes had null values since the cloud screening algorithm used in the Pathfinder processing performed poorly in distinguishing between snow and clouds. The BRDF coefficients of the iterative model and the Fourier coefficients of the temporal model showed a strong spatial structure corresponding to known variations in land cover.

A three–dimensional lattice–gas model for amphiphilic fluid dynamics

We describe a three-dimensional hydrodynamic lattice-gas model of amphiphilic fluids. This model of the non-equilibrium properties of oil-water-surfactant systems, which is a non-trivial extension of an earlier two-dimensional realisation due to Boghosian, Coveney and Emerton [Boghosian, Coveney, and Emerton 1996, Proc. Roy. Soc. A 452, 1221-1250], can be studied effectively only when it is implemented using high-performance computing and visualisation techniques. We describe essential aspects of the model's theoretical basis and computer implementation, and report on the phenomenological properties of the model which confirm that it correctly captures binary oil-water and surfactant-water behaviour, as well as the complex phase behaviour of ternary amphiphilic fluids.

Flow simulation and high performance computing

Flow simulation is a computational tool for exploring science and technology involving flow applications. It can provide cost-effective alternatives or complements to laboratory experiments, field tests and prototyping. Flow simulation relies heavily on high performance computing (HPC). We view HPC as having two major components. One is advanced algorithms capable of accurately simulating complex, real-world problems. The other is advanced computer hardware and networking with sufficient power, memory and bandwidth to execute those simulations. While HPC enables flow simulation, flow simulation motivates development of novel HPC techniques. This paper focuses on demonstrating that flow simulation has come a long way and is being applied to many complex, real-world problems in different fields of engineering and applied sciences, particularly in aerospace engineering and applied fluid mechanics. Flow simulation has come a long way because HPC has come a long way. This paper also provides a brief review of some of the recently-developed HPC methods and tools that has played a major role in bringing flow simulation where it is today. A number of 3D flow simulations are presented in this paper as examples of the level of computational capability reached with recent HPC methods and hardware. These examples are, flow around a fighter aircraft, flow around two trains passing in a tunnel, large ram-air parachutes, flow over hydraulic structures, contaminant dispersion in a model subway station, airflow past an automobile, multiple spheres falling in a liquid-filled tube, and dynamics of a paratrooper jumping from a cargo aircraft.

Computer-integrated mechanical engineering (CIME)

Companies can only survive the rigors of international competition if they can bring telling advantages to bear in markets that are subject to vigorous change. New information technologies hold the key. An unbroken flow of data intermeshing with business management systems can support the development of a product from basic concept to design, manufacture and assembly. As tasks become more complex, the development of mechanical products requires assured mastery of high-performance computational techniques in an environment marked by constantly changing parameters. The potential of new engineering materials, e.g. in the fields of ceramics, powder metallurgy and fiber-reinforced plastics, have to be taken into account. Product quality can be enhanced using integrated computer-aided systems, e.g. for optimizing the geometry of machine components. Machining times and hence costs can be appreciably reduced. Tool optimization is achieved using CAD in conjunction with on-line CAM facilities. To increase availability and enhance product quality and user-friendliness, mechanical engineers will increasingly use computer-aided simulation and diagnosis systems to improve products, methods and techniques. Apart from product quality, a flexible response to customer's wishes and short development and delivery times will be decisive factors in future competitive scenarios. CAD support and the use of expert systems will substantially reduce the lead time from receipt of customer enquiry to submission of an offer. These trends in the development of mechanical products within the framework of an integrated information system will have increasing impact on organizational structures, managerial approaches and personnel requirements.

Cavitation noise as a train of sound pulses generated at random times: Morosov, V.P., Soviet Physics—Acoustics, Vol 14, No 3 (January–March 1969) pp 361–365 (1420)

An object-oriented approach to load-balanced adaptive mesh refinement in two and three dimensions is presented for OpenFOAM, a mature C++ library for computational fluid dynamics. Such a high-performance computing technique is mandatory for computational efficiency in cases of moving regions of interest for distributed-memory parallel computer architectures where domain decomposition is applied. Moving regions of interest are dynamically deforming and migrating through the domain during simulations and require high spatial resolution of solution features. We present software design and code structure, and detail on our achievements with respect to software usability for engineering applications, as well as on numerous developments and improvements necessary for high stability and runtime performance.