Journal of KIISE:Computer Systems and Theory (한국정보과학회논문지:시스템및이론)

Korean Institute of Information Scientists and Engineers (한국정보과학회)
권호 표지

Implementation and Performance Analysis of PC Clusters using Fast PCs& High Speed Network

초고속 네트워크를 이용한 PC 클러스터의 구현과 성능 평가

  • 김영태 (강릉대학교 컴퓨터과학과) ;
  • 이용희 (기상연구소 수치예보실 기상연구사) ;
  • 최준태 (기상연구소 수치예보실 기상연구사) ;
  • 오재호 (부경대학교 환경대기과학과)
  • Published : 2002.02.01
Download PDF
⟨ Previous Next ⟩

Abstract

We implemented two fast PC clusters using fast PCs and high speed network. First. we built the first generation of 16 PC cluster and have used it for real-time operation at Cheju Regional Meteorological Office. Next, we built the second generation of 16PC with dual CUs cluster which was efficiently improved based on performance analysis of the first generation of cluster. In this research we also analyzed performance of two different clusters, which have different CPUs and communication devices using the parallel model MM5 which has been used for the real-time weather forecasting.

본 연구에서는 고속의개인용 컴퓨터와 초고속 통신장비를 이용한 부산/병렬 계산을 위한 클러스터를 구현하였다. 먼저 16강 PC의 1세대의 클러스터를 구현하여 현재 현업(제주 지방기상처)에서 실시 간으로 운영중이며 1세대 클러스터의 성능 분석을 통하여 효율적으로 성능이 개선된 2CPU의 16PC로 구성된 2세대 클러스터를 구현하였다. 본 연구서는 다른 속도의 CPU및 통신 장비로 구성된 두 가지 형태의클러스터를 실제로 기상 예보를 위하여 사용되는 병렬모델인 MM5 를 이용하여 계산 시간 및 통신에 대한 다양한 분석을 하였다.

Keywords

References

  1. Bernaschi, M., and Richelli, G., 'Development and results of PVMe on the IBM 9076 SP1,' Parallel and Distributed Computing, 26, pp. 75-83, 1995 https://doi.org/10.1006/jpdc.1995.1107
  2. Pacheco P., Parallel Programming with MPI, San Francisco, CA: Morgan Kaufmann, 1997
  3. Kim, Y., Pan, Z., Takle, E. and Kothari, S., 'Parallel Implementation of Hydrostatic MM5 (Mesoscale Model),' The 8th SIAM Conference on Parallel Processing for Scientific Computing, Minneapolis USA, Mar. 14-17, 1997
  4. Michalakes, J., Canfielf, T., Nanjundiah, R. and Hammond, S., 'Parallel implementation, validation, and performance of MM5,' Proc. 6th Workshop on the use of Parallel Processors in Meteorology, Reading, U. K., European Center for Medium Range Weather Forecasting, 1994
  5. Grell, G. O., Dudhia, J. and Stauffer, D. R., 'A description of the Fifth-Generation Penn State/NCAR Mesoscale Model (MM5),' Tech Rep., NCAR/TN-398+STR, National Center for Atmosphere Research, Boulder, Colorado, June 1994
  6. Lee, Y., Choi, J., Kim, J. Y. and Kim, Y., 'The Effect of Hi-resolution SST on Storm Scale Prediction in Point of Operational Prediction System,' The Tenth PSU/NCAR Mesoscale Model User's Workshop, Boulder USA. June 21-22, 2000
  7. Kim, Y., Kothari, S., Takle, E. and Pan, Z., 'A run-time library and load balance analysis for parallel atmospheric models.' Symposium on Regional Weather Prediction on Parallel Computer Environments, Athens, Greece, 1997
  8. Foster, I. and Toonen, B., 'Load-Balancing Algorithms for Climate Models,' Proc. Scalable High-Performance Computing Conf. IEEE, pp. 674-681, 1994 https://doi.org/10.1109/SHPCC.1994.296706
  9. Hsieh, J., Leng T., Mashayekhi Y. and Rooholamini R., 'Architectural and Performance Evaluation of GigaNet and Myrinet Interconnects on Clusters of Small-Scale SMP Servers,' Proc. Supercomputing Conference, Dallas, USA, 2000 https://doi.org/10.1109/SC.2000.10027
  10. Lin, W., Lau, R., Hwang, K., Lin, X. and Cheung, P., 'Adaptive Parallel Rendering on Multiprocessors and Workstation Clusters,' IEEE Transactions on Parallel and Distributed Systems, 12:3, pp. 241-258, March 2001 https://doi.org/10.1109/71.914755