Efficient parallel communication schemes for computational fluid dynamics codes

被引:5
|
作者
Gopalaswamy, N
Ecer, A
Akay, HU
Chien, YP
机构
[1] Indiana Univ Purdue Univ, Purdue Sch Engn & Technol, Dept Mech Engn, Computat Fluid Dynam Lab, Indianapolis, IN 46202 USA
[2] Indiana Univ Purdue Univ, Purdue Sch Engn & Technol, Dept Elect Engn, Computat Fluid Dynam Lab, Indianapolis, IN 46202 USA
关键词
D O I
10.2514/2.465
中图分类号
V [航空、航天];
学科分类号
08 ; 0825 ;
摘要
A scheme for improving the efficiency of communications for the parallel computation of Euler equations is presented. The NPARC (National PARC) code is employed as an example for analyzing the flow through a supersonic inlet. The how-held is divided into subregions called blocks. The parallel computation of the problem normally requires communication between each block after each time step of an explicit Runge-Kutta integration scheme. In the developed procedure, the boundary conditions are frozen for k = 10-20 time steps, and blocks are integrated in time without communication with each other during this period. When the boundary conditions are updated, an oscillatory error is introduced into the solution with a fundamental period of 4k time steps, which is then filtered in time, As a result, the communication cost of parallel computing is significantly reduced, Examples for steady and unsteady hows are presented to demonstrate the applicability of the developed procedure.
引用
收藏
页码:961 / 967
页数:7
相关论文
共 50 条
  • [41] The use of proprietary computational fluid dynamics codes for flows in annular packed beds
    Tierney, M
    Nasr, A
    Quarini, G
    SEPARATION AND PURIFICATION TECHNOLOGY, 1998, 13 (02) : 97 - 107
  • [42] DEVELOPMENT OF A PARALLEL COMPUTATIONAL FLUID-DYNAMICS ALGORITHM ON A HYPERCUBE COMPUTER
    BRAATEN, ME
    INTERNATIONAL JOURNAL FOR NUMERICAL METHODS IN FLUIDS, 1991, 12 (10) : 947 - 963
  • [43] A parallel computational code for the eduction of coherent structures of turbulence in fluid dynamics
    Alfonsi, G
    Primavera, L
    PARALLEL COMPUTING TECHNOLOGIES, 2005, 3606 : 381 - 392
  • [44] Pseudospectral element method for computational fluid dynamics and its parallel implementation
    Ku, Hwar-Ching
    Rosenberg, Allan P.
    Johns Hopkins APL Technical Digest (Applied Physics Laboratory), 1991, 12 (03): : 234 - 245
  • [45] Solution of computational fluid dynamics problems on parallel computers with distributed memory
    Chetverushkin, B
    Churbanova, N
    Trapeznikova, M
    INTERNATIONAL CONFERENCE ON PARALLEL PROCESSING WORKSHOPS, PROCEEDINGS, 2001, : 31 - 36
  • [46] Efficient computational fluid dynamics mesh generation by image registration
    Barber, D. C.
    Oubel, E.
    Frangi, A. F.
    Hose, D. R.
    MEDICAL IMAGE ANALYSIS, 2007, 11 (06) : 648 - 662
  • [47] Efficient GPU Implementation of Automatic Differentiation for Computational Fluid Dynamics
    Zubair, Mohammad
    Ranjan, Desh
    Walden, Aaron
    Nastac, Gabriel
    Nielsen, Eric
    Diskin, Boris
    Paterno, Marc
    Jung, Samuel
    Davis, Joshua Hoke
    2023 IEEE 30TH INTERNATIONAL CONFERENCE ON HIGH PERFORMANCE COMPUTING, DATA, AND ANALYTICS, HIPC 2023, 2023, : 377 - 386
  • [48] Schemes for correcting errors in parallel communication
    Zhang, Huan-guo
    Ding, Yu-long
    Qin, Zhong-ping
    Wuhan Daxue Xuebao/Journal of Wuhan University, 2000, 46 (03): : 319 - 321
  • [49] Computational Fluid Dynamics
    Wagner, Siegfried
    HIGH PERFORMANCE COMPUTING IN SCIENCE AND ENGINEERING '08, 2009, : 167 - 169
  • [50] Computational fluid dynamics
    Wagner, S
    HIGH PERFORMANCE COMPUTING IN SCIENCE AND ENGINEERING 01, 2002, : 269 - 271