We developed a parallel computing method for rapid eddy current analyses using the A - phi method. In this paper, the outline of the developed method is described. Moreover, the performance of the proposed method runn...
详细信息
We developed a parallel computing method for rapid eddy current analyses using the A - phi method. In this paper, the outline of the developed method is described. Moreover, the performance of the proposed method running on a PC cluster is quantitatively clarified through the analyses of some rotating machines.
This paper deals with the discretization of the problem of mould filling in iron foundry and its numerical solution using a Schwarz domain decomposition method. An adapted technique for domain decomposition methods th...
详细信息
This paper deals with the discretization of the problem of mould filling in iron foundry and its numerical solution using a Schwarz domain decomposition method. An adapted technique for domain decomposition methods that suits the discretization in time by the method of characteristics is introduced. Furthermore, the projection method is used to reduce the computation time. Finally, numerical experiments show and validate the effectiveness of the proposed scheme.
Fully discretized models for image reconstruction from projections give rise to huge and sparse nonlinear optimization problems. We study the use of parallel and vector supercomputers for the iterative reconstruction ...
详细信息
Fully discretized models for image reconstruction from projections give rise to huge and sparse nonlinear optimization problems. We study the use of parallel and vector supercomputers for the iterative reconstruction of medical images. A block-iterative version of the Multiplicative Algebraic Reconstruction Technique (MART) is implemented on a CRAY X-MP/48. The implementation exploits the block structure of the algorithm which allows us to take advantage both of the vector architecture of the computer and of the multiple processors for parallel computations. Results indicate that block-iterative algorithms are suitable for parallel and vector implementations and can reconstruct with relative efficiency highly discretized images which give rise to very large optimization problems.
Continuous genetic evaluation of dairy cattle with test-day models is desired in Finland. However, the computing time for the genetic evaluation is 4 d and exceeds the minimum of a weekend. Three parallel implementati...
详细信息
Continuous genetic evaluation of dairy cattle with test-day models is desired in Finland. However, the computing time for the genetic evaluation is 4 d and exceeds the minimum of a weekend. Three parallel implementations of the preconditioned conjugate gradient iterative solver were programmed and compared to identify the best strategy for solving mixed model equations using parallel computing. The programs were used to solve two random regression test-day models with approximately 7.28 and 49.9 million unknowns. The latter model will be used in the Finnish dairy cattle evaluation. computing times for the smaller model with the four processors available were 52, 32, and 27% of the single processor program when the complexity of the parallel program was increased. In practice, the best program required the most programming because the other parallel programs could not solve the larger model because of excess memory requirements. parallel computing with four processors reduced the time to obtain solutions of Finnish dairy cattle evaluations to under 2 d. Benefit from parallel computing will be increased if amount of computing memory is increased.
This paper proposes two viable computing strategies for distributed parallel systems: domain division with sub-domain overlapping and asynchronous communication. We have implemented a parallel computing procedure for ...
详细信息
This paper proposes two viable computing strategies for distributed parallel systems: domain division with sub-domain overlapping and asynchronous communication. We have implemented a parallel computing procedure for simulation of Ti thin film growing process of a system with 1000 x 1000 atoms by means of the Monte Carlo (MC) method. This approach greatly reduces the computation time for simulation of large-scale thin film growth under realistic deposition rates. The multi-lattice MC model of deposition comprises two basic events: deposition, and surface diffusion. Since diffusion constitutes more than 90% of the total simulation time of the whole deposition process at high temperature, we concentrated on implementing a new parallel diffusion simulation that reduces communication time during simulation. Asynchronous communication and domain overlapping techniques are used to reduce the waiting time and communication time among parallel processors. The parallel algorithms we propose can simulate the thin film growth of a system with many more particles than before under realistic deposition rates, and can provide a more efficient means for computer simulation of thin film growth.
This paper describes the application of parallel computing technologies in systems with shared and distributed memory for solving optimization problems of geometric design. The first technology is based on the maximin...
详细信息
This paper describes the application of parallel computing technologies in systems with shared and distributed memory for solving optimization problems of geometric design. The first technology is based on the maximin properties of phi-functions for composite objects, and the second technology uses the multistart strategy and methods for minimizing nonsmooth functions. This allowed to several times reduce time expenditures for searching for locally optimum placements of 2D and 3D objects and to obtain better results as for the objective function value.
This paper discusses the parallel computing of the third generation Ocean General Circulation Model (OGCM) from the State Key Laboratory of Numerical Modeling for Atmospheric Science and Geophysical Fluid Dynamics(LAS...
详细信息
This paper discusses the parallel computing of the third generation Ocean General Circulation Model (OGCM) from the State Key Laboratory of Numerical Modeling for Atmospheric Science and Geophysical Fluid Dynamics(LASG),Institute of Atmosphere Physics(IAP). Meanwhile, several optimization strategies for parallel computing of OGCM (POGCM) on Scalable Shared Memory Multiprocessor (S2MP) are presented. Using Message Passing Interface (MPI), we obtain super linear speedup on SGI Origin 2000 for parallel OGCM(POGCM) after optimization.
To deal with the high dimensionality and computational density of the Optimal Power Flow model with Transient Stability Constraints(OTS),a credible criterion to determine transient stability is proposed based on swing...
详细信息
To deal with the high dimensionality and computational density of the Optimal Power Flow model with Transient Stability Constraints(OTS),a credible criterion to determine transient stability is proposed based on swing curves of generator rotor and the characteristics of transient *** this method,the swing curves of all generator rotors will be independent one ***,when a parallel computing approach based on the MATLAB parallel toolbox is used to handle multi-contingency cases,the calculation speed is improved ***,numerical simulations on three test systems including the NE-39 system,the IEEE 300-bus system,and 703-bus systems,show the effectiveness of the proposed method in reducing the computing time of OTS calculation.
We develop an algorithm to calculate invariant distributions of large Markov chains whose state spaces are partitioned into "islands" and "ports". An island is a group of states (cluster) with pote...
详细信息
We develop an algorithm to calculate invariant distributions of large Markov chains whose state spaces are partitioned into "islands" and "ports". An island is a group of states (cluster) with potentially many connections inside of the island but a relatively small number of connections between islands. The states connecting different islands are called ports. Our algorithm is developed in the framework of the "state reduction approach", but the special structure of the state space allows calculation of the invariant distribution to be done in parallel. Additional problems such as computation of fundamental matrices and optimal stopping problems are also analyzed for such Markov chains.
In recent years, Grid computing evolved from first implementations as prototype Grid environments to large-scale production Grid infrastructures utilised during everyday work by scientists around the world. This demon...
详细信息
In recent years, Grid computing evolved from first implementations as prototype Grid environments to large-scale production Grid infrastructures utilised during everyday work by scientists around the world. This demonstrates that the concept of the Grid is more than merely a marketing phrase, but instead an enabler for new application domains in parallel and distributed computing. Among others, the EU project EGEE ("Enabling Grids for E-Science") is probably the worlds largest initiative for establishing a permanent Grid infrastructure on a 24 x 7 basis. Such a Grid that is always on and there to serve the community just like the ubiquitous networking infrastructures today is on the horizon, waiting for users to utilise it in their applications. This paper explores the different possibilities for utilising Grids in parallel applications, focusing on different parallel computing aspects as provided by Grid environments. Basic concepts such as using the Grid as a large data storage and management basis or as a collection of distributed supercomputers represent a new approach to computational science, where users are expecting to utilise unprecedented amounts of performance for tackling their scientific problems.
暂无评论