Performance Analysis of a Linux Pc Cluster Using a Direct Numerical Simulation of Fluid Turbulence Code

Abstract

In this paper, we have tested the parallel performance of an Intel Xeon-based Linux PC cluster using a finite element code for direct numerical simulation (DNS) of incompressible fluid turbulence. The parallel performance of the PC cluster, which used up to 64 2.8 GHz processors, was evaluated by comparing three scales of DNS trial runs consisting of 3.3, 5.8, and 10.1 million elements. Subroutines of different natures were contrasted to investigate the scalability of the DNS code. For DNS calculation of sufficiently large scale, the subroutines showed reasonable parallel efficiency. Doubling the number of processors reduced the CPU time by about 40%. Of particular interest was the CPU time required by the two subroutines handling interprocessor communication that was fairly constant within the range of processors tested. PC Linux clusters are thus affordable platforms, compared with more expensive supercomputers, to conduct large-scale scientific computing for fluid turbulence research.

Keywords

Computational fluid dynamics direct numerical simulation fluid turbulence Linux cluster distributed and parallel computing finite element method

Get full access to this article

View all access options for this article.

References

Barrett, R. , Berry, F. , Chan, T. F. , Demmel, J. , Donato, J. Dongarra, J. , Eijkhout, V. , Pozo, R. , Romine, C. , and Van der Vorst, H. 1994. Templates for the Solution of Linear Systems: Building Blocks for Iterative Methods, 2nd edition. SIAM, Philadelphia, PA .

Chernyavsky, B. and Knight, D. 2001. Investigation of largeeddy simulation code scaling performance and network type influence on a linux PC cluster . International Journal of High Performance Computing Applications 15: 388–393 .

Garbey, M. and Vassilevski, Y. V. 2001. A parallel solver for unsteady incompressible 3D Navier–Stokes equations . Parallel Computing 27: 363–389 .

Johan, Z. , Hughes, T. J. R. , Mathur, K. K. , and Johnsson, S. L. 1992. A data parallel finite element method for computational fluid dynamics on the Connection Machine System . Computer Methods in Applied Mechanics and Engineering 99: 113–134 .

Liu, C-H. 2003. Turbulent plane Couette flow and scalar transport at low Reynolds number . Transactions of the ASME, Journal of Heat Transfer 125: 988–998 .

Liu, C-H. and Barth, M. C. 2002. Large-eddy simulation of flow and scalar transport in a modeled street canyon . Journal of Applied Meteorology 41: 660–673 .

Liu, C-H. and Leung, D. Y. C. 1998. Parallel computation of atmospheric pollutant dispersion under unstably stratified atmosphere . International Journal for Numerical Methods in Fluids 26: 677–696 .

Liu, C-H. and Leung, D. Y. C. 2001. Development of a finite element solution for the unsteady Navier–Stokes equations using projection method and fractionalscheme . Computer Methods in Applied Mechanics and Engineering 190: 4301–4317 .

Liu, C-H. , Leung, D. Y. C. , and Woo, C-M. 2003. Development of a scalable finite element solution to the Navier–Stokes equations . Computational Mechanics 2: 490–498 .

10.

Lundin, L. K. 1998. Computing the velocity of a rotating flow . Parallel Computing 24: 2021–2034 .

11.

Passoni, G. , Alfonsi, G. , Tula, G. , and Cardu, U. 1999. A wave number parallel computational code for the numerical integration of the Navier–Stokes equations . Parallel Computing 25: 593–611 .

12.

Passoni, G. , Cremonesi, P. , and Alfonsi, G. 2001. Analysis and implementation of a parallelization strategy on a Navier–Stokes solver for shear flow simulations . Parallel Computing 27: 1665–1685 .

13.

Rathish Kumar, B. V. , Yamaguchi, T. , Liu, H. , and Himeno, R. A parallel 3D unsteady incompressible flow solver on VPP700 . Parallel Computing 27: 1687–1713 .

14.

Sterling, T. 2001. An introduction to PC clusters for high performance computing . International Journal of High Performance Computing Applications 15: 92–101 .

15.

Wasfy, T. , West, A. C. , and Modi, V. 1998. Parallel finite element computation of unsteady incompressible flows . International Journal for Numerical Methods in Fluids 26: 17–37 .