• Journal of Korea Water Resources Association
• /
• v.43 no.8
• /
• pp.747-755
• /
• 2010

Distributed Models have relative weak points due to the amount of computer memory and calculation time required for calculating water flow using a numerical analysis based on kinematic wave theory when compared to the conceptual models used so far. Typically, the distributed models have been mainly applied to small basins. It was necessary to decrease the resolution of the grid to make it applicable for large scale watersheds, and because it would take up too much time to calculate using a higher resolution. That has been one of the more difficult factors in applying the model for actual work. In this paper, MPI (Message Passing Interface) technique was applied to solve the problem of calculation time as it is one of the demerits of the distributed model for performing physical and complicated numerical calculations for large scale watersheds. The comparison studies were performed a single domain and a divided small domain in Yongdam Dam watershed in case of typoon 'Ewiniar' at 2006. They were compared to analyze the application effects of parallelization technique. As a result, a maximum of 10 times the amount of calculation time was saved but keeping the level of quality for discharge by using parallelization code rather than a single processor.

• Journal of KIISE
• /
• v.42 no.12
• /
• pp.1467-1473
• /
• 2015

In this paper, we implemented a parallel random number generation program on GPU's, which are known for high performance computing, using LCG (Linear Congruential Generator). Random numbers are important in all fields requiring the use of randomness, and LCG is one of the most widely used methods for the generation of pseudo-random numbers. We explained the parallel program using the NVIDIA CUDA model and MPI(Message Passing Interface) and showed uniform distribution and performance results. We also used a Monte Carlo algorithm to calculate pi(${\pi}$) comparing the parallel random number generator with cuRAND, which is a CUDA library function, and showed that our program is much more efficient. Finally we compared performance results using multi-GPU's with those of ideal speedups.

• 한국전산유체공학회:학술대회논문집
• /
• 2004.10a
• /
• pp.111-114
• /
• 2004

CFD code that simulates stator-rotor interactions is developed applying parallel computing method. Modified Multi-Block Grid System which enhances perpendicularity in grid and is appropriate in parallel processing is introduced and Patched Algorithm is applied in sliding interface which is caused by movement of rotor. The experimental model in the turbo-machine is composed of 11 stators and 14 rotors. Analyses on two test cases which are one stator - one rotor model and three stators - four rotors model are performed. The results of the two cases have been compared with the experimental test data.

• Proceedings of the KSME Conference
• /
• 2002.08a
• /
• pp.537-540
• /
• 2002

Flow through turbomachinery has a very complex structure and is intrinsically unsteady. Especially, recent design trend to turbomachinery with short axial spacing makes the flow extremely complex due to the interaction between stator and rotor. Therefore, it is very necessary to clearly understand the complex flow structure to obtain the high efficiency turbomachinery. So, in this paper, the effects of axial spacing on the unsteady secondary flow performance in the one stage turbine are investigated by three-dimensional unsteady flow analysis. The three-dimensional solver is parallelized using domain decomposition and Message Passing Interface(MPI) standard to overcome the limitation of memory and the CPU time in three-dimensional unsteady calculation. A sliding mesh interface approach has been implemented to exchange flow information between blade rows.

• Proceedings of the KSME Conference
• /
• 2000.04b
• /
• pp.586-592
• /
• 2000

터보머신 태부에 존재하는 정익 - 동익의 상호작용 유동현상을 수치모사 하는 코드를 병렬화 하였다 정익 - 동익의 상호작용을 해석하는 데에 편리하도륵 Multi-Block Grid System을 도입하여 계산영역을 형성하였고, 동익의 움직임으로 인해 발생하는 Sliding Interface부분은 Patched 알고리즘을 적용하여 해석하였다. 정익과 동익의 수를 1대 1로 단순화시켜 수치모사한 결과와 정익과 동익의 수를 실제 조건과 더 비슷하게 설정한 3대 4의 비율로 맞추어 수치모사한 결과를 비교하였다. 또한, 병렬컴퓨팅으로 인해 단축된 계산시간을 다른 연구에서의 계산시간들과 서로 비교하였다. 2차원 비정상 압축성 Navier-Stokes 방정식이 이용되었고, 난류모델링에는 K-w SST 모델링이 적응되었다. Roe의 FDS 기법을 사용하여 플럭스를 계산하였고, MUSCL 기법을 적용하여 3차의 공간정확도를 갖도록 하였다. 시간적분에는 이보성의 DP-SGS를 사용하였다. 해석결과의 분석에는 Time-averaged pressure distribution과 Pressure amplitude distribution 데이터를 사용했다.

• Journal of Korean Society for Geospatial Information Science
• /
• v.18 no.4
• /
• pp.21-30
• /
• 2010

In this paper, the parallel distributed rainfall runoff model(K-DRUM) using MPI(Message Passing Interface) technique was developed to solve the problem of calculation time as it is one of the demerits of the distributed model for performing physical and complicated numerical calculations for large scale watersheds. The K-DRUM model which is based on GIS can simulate temporal and spatial distribution of surface flow and sub-surface flow during flood period, and input parameters of ASCII format as pre-process can be extracted using ArcView. The comparison studies were performed with various domain divisions in Namgang Dam watershed in case of typoon 'Ewiniar' at 2006. The numerical simulation using the cluster system was performed to check a parallelization effectiveness increasing the domain divisions from 1 to 25. As a result, the computer memory size reduced and the calculation time was decreased with increase of divided domains. And also, the tool was suggested in order to decreasing the discharge error on each domain connections. The result shows that the calculation and communication times in each domain have to repeats three times at each time steps in order to minimization of discharge error.

• Journal of Digital Contents Society
• /
• v.19 no.2
• /
• pp.343-349
• /
• 2018

The proposed hybrid algorithm combines the benefits of rapid convergence property of mean filed annealing(MFA) and the effective genetic operations of simulated annealing-like genetic algorithm(SGA). This algorithm is applied to the isotropic material stock cutting problem, especially to glass cutting in distributed computing environments base on MPI called message passing interface. The glass cutting is to place the required rectangular patterns to the given large glass sheets resulting in reducing the wasted scrap area. Our experimental results show that the heuristic method improves the performance over the conventional ones by decreasing the scrap area and maximum execution time. It is also proved that the proposed distributed algorithm maintains the convergence properties of sequential one while it achieves almost linear speedup as the problem size increases.

• 한국신재생에너지학회:학술대회논문집
• /
• 2006.06a
• /
• pp.382-385
• /
• 2006

한국지질자원연구원은 1997년부터 새로운 에너지 자원으로 활용 가능성을 포함하고 있는 가스 하이드레이트를 조사하기 위해 동해 일원에서 탄성파탐사를 실시하고 있다. 탄성파 반사 자료로부터 가스 하이드레이트 부존여부를 확인하는 방법은 해저면과 평행하면서 위상이 반대로 나타나는 고진폭 반사파 BSR (Bottom Simulating Reflection)과 BSR상부에서의 진폭감소, 하부에서 진폭증가와 구간속도 감소 등을 들 수 있다. 여기에서는 가스 하이드레이트 탐사자료에 대한 일반자료처리와 함께 BSR을 포함하고 있는 탄성파 반사자료에 대해 코드 병렬화된 PSPI를 이용하여 깊이영역 구조보정을 실시하였다. 고용량 탐사자료로 구성된 탄성파 반사자료에 깊이영역 구조보정을 적용하기 위해서는 고성능 컴퓨터와 병렬처리 기술이 필요하다. PSPI(Phase Shift Plus Interpolation)법은 적은 컴퓨터 계산량과 효율성 그리고 주파수 영역에서 구조적으로 병렬화가 용이한 특성을 지니고 있어 구조보정에 많이 이용되고 있다. 여기에서는 MPI(Message Passing Interface)-LAM을 이용하여 병렬코드화된 PSPI를 개발하고 인공합성모델과 동해 가스 하이드레이트 깊이영역 구조보정에 적응하였다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2013.05a
• /
• pp.476-477
• /
• 2013

Gadget-2 is a scientific simulation code has been used for many different types of simulations like, Colliding Galaxies, Cluster Formation and the popular Millennium Simulation. The code is parallelized with Message Passing Interface (MPI) and is written in C language. There is also a Java adaptation of the original code written using MPJ Express called Java Gadget. Java Gadget writes a lot of checkpoint data which may or may not use the HDF-5 file format. Since, HDF-5 is MPI-IO compliant, we can use our MPJ-IO library to perform parallel reading and writing of the checkpoint files and improve I/O performance. Additionally, to add reliability to the code execution, we propose the usage of Hadoop Distributed File System (HDFS) for writing the intermediate (checkpoint files) and final data (output files). The current code writes and reads the input, output and checkpoint files sequentially which can easily become bottleneck for large scale simulations. In this paper, we propose Sim-Hadoop, a framework to leverage HDFS and MPJ-IO for improving the I/O performance of Java Gadget code.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.45 no.11
• /
• pp.932-938
• /
• 2017

Particle-in-cell method which blends Eulerian grids and Lagrangian particle is utilized to solve simplified hall-effect thruster. Since this study individually tracks not only neutrons and ions but also electrons, message passing interface(mpi) scheme is adopted for parallel computer cluster. Helical movement of an electron cloud in constant magnetic field is validated comparing with an exact solution. A plasma in radial magnetic field and axial electric field in a reaction cylinder is established. Electrons do double helix movement and are well anchored in a cylinder. Ionization of neutrons by impact with high-speed electrons generates ion particles. They are accelerated by axial electric field, which forms a plume of a plasma-effect thruster.