• 한국항공우주학회지
• /
• 제30권6호
• /
• pp.21-30
• /
• 2002

본 연구에서는 영역분할기법에 의하여 예조건화 Navier-Stokes 방정식을 병렬화 하였으며, 병렬화 된 코드의 정확도는 순차 코드의 결과 및 실험 데이터와의 비교를 통하여 확인하였다. 코드의 병렬효율은 Myrinet을 기반의 PC 클러스터와 Fast-Ethernet PC 클러스터에서 살펴보았다. 주된 성능 지표로는 프로세서 수와 네트웍 통신 구성에 따른 속도 향상 비를 살펴보았다. 이 시험에서 Myrinet 환경의 PC 클러스터는 기대한 바와 같이 Fast-Ethernet에 비하여 우수한 성능을 보여 주었다. 문제의 크기에 대한 의존도 시험에서 네트웍 통신 속도는 병렬처리 성능에 중요한 요소이며, Myrinet 기반의 PC 클러스터가 고성능 병렬처리 시스템의 한 가지 대안임을 보여 주었다.

• 지구물리와물리탐사
• /
• 제19권3호
• /
• pp.131-135
• /
• 2016

코어레이는 포트란 2008 표준에 도입된 병렬 연산 기법이다. 코어레이를 이용하면 간단한 문법으로 분산 메모리시스템에서 병렬 연산을 구현할 수 있다. 본 연구에서는 탄성파 자료 처리 프로그램에 코어레이와 MPI를 적용하여 병렬 처리 성능을 비교하고 이를 통해 코어레이의 적용 가능성을 살펴보았다. 파동 전파 모델링을 이용해 연산 성능을 비교하였고, 영역 분해 기법을 이용해 일대일 통신 성능을 비교하였다. 또한 거꿀 참 반사 보정 프로그램을 이용해 병렬 처리 성능을 비교하였다. 그 결과 연산 성능은 코어레이 프로그램과 MPI 프로그램에서 큰 차이가 없었지만 통신 성능은 MPI가 우수했다.

• 한국전산유체공학회:학술대회논문집
• /
• 한국전산유체공학회 2003년도 The Fifth Asian Computational Fluid Dynamics Conference
• /
• pp.84-85
• /
• 2003

A parallelized FEM code based on domain decomposition method has been recently developed for a large scale computational fluid dynamics. A 4-step splitting finite element algorithm is adopted for unsteady computation of the incompressible Navier-Stokes equation, and Smagorinsky LES(Large Eddy Simulation) model is chosen for turbulent flow computation. Both METIS and MPI library are used for domain partitioning and data communication between processors respectively. Tiburon of Hyundai-motor is chosen as the computational model at $Re=7.5{\times}10^{5}$, which is based on the car height. It is confirmed that the drag under road condition is smaller than that of wind tunnel condition.

• Journal of Mechanical Science and Technology
• /
• 제19권4호
• /
• pp.1006-1017
• /
• 2005

In this work, we developed an accurate and efficient radiative finite volume method applicable for the complex 2D planar and 3D geometries using an unstructured-grid finite volume method. The present numerical model has fully been validated by several benchmark cases including the radiative heat transfer in quadrilateral enclosure with isothermal medium, tetrahedral enclosure, a three-dimensional idealized furnace, as well as convection-coupled radiative heat transfer in a square enclosure. The numerical results for all cases are well agreed with the previous results. Special emphasis is given to the parallelization of the unstructured-grid radiative FVM using the domain decomposition approach. Numerical results indicate that the present parallel unstruc­tured-grid FVM has the good performance in terms of accuracy, geometric flexibility, and computational efficiency.

• Journal of Mechanical Science and Technology
• /
• 제17권12호
• /
• pp.1886-1893
• /
• 2003

This paper examines the parameter observability of a calibration system that consrains a mobile platform to a planar table to take the calibration data. To improve the parameter observability, we find the optimal configurations providing the calibration with maximum contribution. The QR-decomposition is used to compute the optimal configurations that maximize the linear independence of rows of an observation matrix. The calibration system is applied to the parallel type manipulator constructed for a machining center. The calibration results show that all the necessary kinematic parameters assigned in a Stewart-Gough platform are identifiable and convergent to desirable accuracy.

• 한국전산유체공학회:학술대회논문집
• /
• 한국전산유체공학회 2005년도 춘계 학술대회논문집
• /
• pp.123-127
• /
• 2005

The numerical analyses of the complicated flows are widely attempted in these days. Because of the enormous demanding memory and calculation time, parallel processing is used for these problems. In order to obtain calculation efficiency, it is important to choose proper domain decomposition technique and numerical algorithm. In this research we enhanced the efficiency of the CFDS code developed by ADD, using parallel computation and newly developed numerical algorithms. For the huge amount of data transfer between blocks non-blocking method is used, and newly developed data transfer algorithm is used for non-aligned block interface. Recently developed RoeM scheme is adpoted as a spatial difference method, and AF-ADI and LU-SGS methods are used as a time integration method to enhance the convergence of the code. Analyses of the flows around the ONERA M6 wing and the high angle of attack missile configuration are performed to show the efficiency improvement.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권11호
• /
• pp.5287-5303
• /
• 2018

Handling vast amount of data found in large transactional datasets is an obvious challenge for the conventional data mining algorithms. Addressing this challenge, our paper proposes a parallel approach for proper decomposition of mining problem into sub-problems in order to find frequent patterns from these datasets. The proposed, Pattern Mining for Large Distributed Dataset (PMLDD) approach, ensures minimum dependencies as well as minimum communications among sub-problems. It establishes a linear aggregation of the intermediate results so that it can be adapted to large-scale programming models like MapReduce. In this context, an algorithmic structure for MapReduce programming model is presented. PMLDD guarantees an efficient load balancing among the sub-problems by a specific selection criterion. Further, it optimizes the number of required iterations over the dataset for mining frequent patterns as compared to the existing approaches. Finally, we believe that our approach is scalable enough to handle larger datasets in terms of performance evaluation, and the result analysis justifies all these mentioned concerns.

• Journal of Electrical Engineering and Technology
• /
• 제2권3호
• /
• pp.335-341
• /
• 2007

In this study, we present an approach to parallelizing OPF that is suitable for distributed implementation and is applicable to very large inter-connected power systems. The approach could be used by utilities for optimal economy interchange without disclosing details of their operating costs to competitors. It could also be used to solve several other computational tasks, such as state estimation and power flow, in a distributed manner. The proposed algorithm was demonstrated with several case study systems.

• 대한수학회논문집
• /
• 제11권2호
• /
• pp.495-501
• /
• 1996

There is currently a regain of interest in ADI (Alternating Direction Implicit) method as a preconditioner for iterative Method for solving large sparse linear systems, because of its suitability for parallel computation. However the classical ADI is not applicable to FE(Finite Element) matrices. In this paper wer propose a Block-ADI method, which is applicable to Finite Element metrices. The new approach is a combination of classical ADI method and domain decompositi on. Also, we provide a partial proof of the convergence based on the results from the regular splittings, in case the discretization metrix is symmetric positive definite.

• 한국전산구조공학회:학술대회논문집
• /
• 한국전산구조공학회 2006년도 정기 학술대회 논문집
• /
• pp.291-298
• /
• 2006

A parallelized topology design optimization method is developed on a distributed memory system. The parallelization is based on a domain decomposition method and a boundary communication scheme. For the finite element analysis of structural responses and design sensitivities, the PCG method based on a Krylov iterative scheme is employed. Also a parallelized optimization method of optimality criteria is used to solve large-scale topology optimization problems. Through several numerical examples, the developed method shows efficient and acceptable topology optimization results for the large-scale problems.