簡易檢索 / 詳目顯示

回結果列表
研究生： 馬康彬
Ma, Kang-Bin
論文名稱： 使用英特爾Xeon PHI協處理器加速模擬使用非結構四面體網格之暫態可壓縮流
Accelerated Transient Compressible Flow Simulations using Unstructured Tetrahedral Grids on the Intel Xeon Phi Coprocessor
指導教授： 李汶樺
Matt-Hew Smith
學位類別： 碩士
Master
系所名稱： 工學院 - 機械工程學系
Department of Mechanical Engineering
論文出版年： 2015
畢業學年度： 103
語文別： 英文
論文頁數： 162
中文關鍵詞： 平行運算 、計算流體力學 、英特爾Xeon PHI協處理器 、開放式多處理 、有限體積法 、精確黎曼解
外文關鍵詞： Parallel Computing, Computational Fluid Dynamics, Intel Xeon PHI Coprocessor, OpenMP, Finite Volume Method, Exact Riemann Solver, Quiet Direct Simulation
相關次數： 點閱：146  下載：1 
分享至:
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報

    • 傳統的計算流體力學(CFD)在三維流場模擬時，為了達成一定的準確性，網格數量比起過去的模擬有越來越多的趨勢，因此需要相當大量的計算以達到需求。為了減少計算流體力學模擬所需的時間，許多研究人員使用平行運算 - 利用多個中央處理器同時運算，有效的分配計算量。然而，典型用來平行計算流體力學的高速電腦(HPC)叢集相當的昂貴，使得此類平行運算方式對於小規模的公司來說較難實行。
    近來英特爾公司研究發展出Xeon PHI協處理器並已上市，此設備包含了許多中央處理器在內，此協處理器可以安置在一般的電腦系統，以增進運算的能力。本研究將呈現暫態、可壓縮計算流體力學解法之研究及應用，並使用Xeon PHI協處理器的運算能力利用非結構四面體網格模擬流體流經複雜的三維幾何環境。本論文之解法 - 精確黎曼解(exact Riemann solver)以及QDS解對於此流體的運算速度範圍大約為傳統Xeon中央處理器運算速度的10-15倍，同時花費大約只需傳統高速電腦叢集的五分之一。
    本研究將針對英特爾Xeon PHI協處理器的多核心特性，利用開放式多處理(OpenMP)的方式，使用適當的執行緒數量及分配方式，達成設備之最佳使用。研究結果顯示於許多業界實際應用，運算效能將於本研究中詳細展示。

    Conventional transient Computational Fluid Dynamics (CFD) simulations are very computationally intensive for three dimensional flows. Since CFD simulation nowadays needs a larger computational grid in order to get a more precise solution, as such it needs more computational time to reach a reasonable result. In order to reduce the time required for CFD computation, many researchers have employed parallel computing - the use of multiple CPU cores, cooperating to effectively share the workload. However, typical High Performance Computing (HPC) clusters used for parallel CFD are very expensive, making this alternative unavailable for many small engineering companies. Recent developments by Intel have resulted in the release of the Xeon Phi coprocessor - a device containing a large number of CPU cores - which can be added to a conventional computer system to increase the computational capability. This research presents the development and application of a transient, compressible CFD solver using an unstructured tetrahedral grid to simulate three dimensional flows through complex geometries by using the computational power of the Xeon Phi coprocessor. The resulting solvers - an exact Riemann solver and a QDS solver - are capable of computing flows at speeds equivalent to approximately 10-15 conventional Xeon CPU cores while only costing approximately 1/5th (20 percent) that of a conventional HPC workstation. This research will cover the performance characteristics of the Many Integrated Core (MIC) Architecture of Xeon PHI coprocessor.

    In addition to the application of an exact Riemann solver to Phi parallelization, this research has applied the Quiet Direct Simulation (QDS) solver to unstructured parallel computation. Details of the implementation are described within, and results are shown for several industrial applications. The performance characteristics of QDS compared to the analytical Riemann solver are described in detail.

    摘要 i Abstract iii Acknowledgements v Contents vi List of Table viii List of Figures x Nomenclature xiv Chapter 1 - Introduction 1 1.1 Background and Motivation 1 1.2 Computational Fluid Dynamics 2 1.3 Governing Equations 3 1.3.1 Continuity Equation 3 1.3.2 Navier-Stokes and Euler Equations 4 1.4 Finite Volume Methods 5 1.4.1 CFL number 8 1.5 Riemann Solvers 9 1.5.1 Elementary Wave Solutions of the Riemann Problem 10 1.5.2 Analytical Solution for the Riemann Problem 14 1.6 Equilibrium Flux Method 15 1.7 Quiet Direct Simulation method 17 1.8 Parallel Computing 19 1.8.1 Parallel Computing Theory 20 1.8.2 Open Multi-Processing (OpenMP) 24 1.8.3 Graphics Processing Unit (GPU) 25 1.9 CPU Architecture 26 1.10 Intel Xeon PHI Coprocessor 27 1.10.1 Background of the Intel Xeon PHI 28 1.10.2 Intel Many Integrated Core (MIC) Architecture 29 1.10.3 Thread Affinity Control 31 1.10.4 Monitoring the Intel Xeon PHI 35 Chapter 2 – Methodology 36 2.1 Preprocessing 36 2.2 Implementation 38 2.2.1 Finite Volume Method Implementation 38 2.2.2 Riemann Solver Implementation 40 2.2.3 Quiet Direct Simulation Implementation 46 2.3 OpenMP Parallelization 48 2.4 Intel Xeon PHI Parallelization 50 2.4.1 Programing Using Intel Xeon PHI 50 2.4.2 Compiler Instructions 52 2.4.3 Executing PHI Programs on Linux Operating Systems 53 Chapter 3 – Numerical Results and Parallel Performance 54 3.1 One-dimensional Shock Tube Problem 55 3.2 Numerical Diffusion Test 57 3.3 Forward Facing Step Problem 58 3.4 Blast Wave Problem 60 3.5 Analysis of Parallel Performance 62 3.5.1 Compiler Optimization Performance 62 3.5.2 OpenMP Parallel Performance 64 3.5.3 Parallelized PHI Thread Performance 65 3.5.4 PHI Thread Affinity Performance 66 3.5.5 Problem Size Performance Using PHI 67 Chapter 4 – Conclusion 69 References 71 Tables 75 Figures 92

    [1] Albright BJ, Daughton W, Lemons DS, Winske D, Jones ME., “Quiet direct simulation of plasmas”, Phys. Plasmas 2002.
    [2] Albright BJ, Lemons DS, Jones ME, Winske D, “Quiet direct simulation of Eulerian fluids”, Phys. Rev E, 2002.
    [3] Amdahl. G.M., “Validity of the single processor approach to achieving large scale computing capabilities”, In AFIPS Conference Proceedings Vol. 30, pp. 483–485, 1967.
    [4] Bhaskaran, G., Gaurav, P., “Optimizing Performance of ROMS on Intel Xeon Phi”, International Conference On Computational Science, Volume 51, Pages 2854–2858, 2015.
    [5] Bird, G.A., “Molecular Gas Dynamics and the direct simulation of gas flows”, Clarendon Press, Oxford, 1994.
    [6] Cebeci, T., Shao, J.P., Kafyeke, F. and Laurendeau, E., “Computational Fluid Dynamics for Engineers”, Horizons Publishing Inc., California, 2005.
    [7] Cramer, T., Dirk Schmidl, Michael Klemm, Dieter an Mey, “OpenMP Programming on Intel Xeon Phi Coprocessors: An Early Performance Comparison”, Center for Computing and Communication.
    [8] Emery, A. F. “J. Comp. Phys.”, 1968.
    [9] Ferziger, Joel H., Peric, Milovan, “Computational Methods for Fluid Dynamics”, Springer-Verlag Berlin Heidelberg, 2002.
    [10] Flynn. M. J., “Very High-Speed Computing Systems”, Proc. IEEE, Vol. 54, pp. 1901-1909, 1966.
    [11] Flynn's taxonomy. (n.d.). Retrieved April 5, 2015, from http://en.wikipedia.org/wiki/Flynn's_taxonomy.
    [12] Gentry, R. A., Martin, R. E., Daly, J. B., “An Eulerian differencing method for unsteady compressible flow problems”, Journal of Computational Physics 1 (1): 87–118, 1966.
    [13] Harlow, F.H., “A Machine Calculation Method for Hydrodynamic Problems”, Los Alamos Scientific Laboratory report LAMS-1956, 1955.
    [14] Hirsch, C., “Numerical Computation of Internal and External Flows”, Vol. 1, John Wiley and Sons, New York, 1998.
    [15] Intel® Xeon Phi™ Coprocessor - the Architecture. 2012. Retrieved May 10, 2015, from https://software.intel.com/en-us/articles/intel-xeon-phi-coprocessor-codename-knights-corner.
    [16] J. Robinet, J. dressier, G. Casalis and J. Moschetta, “Shock wave instability and the carbuncle phenomenon: Same intrinsic origin?”, J. Fluid Mech. (2000), vol. 417, pp. 237–263, 2000.
    [17] Jameson, A., Schmidt, W. and Turkel, E., “Numerical Solution of the Euler Equations by Finite Volume Methods Using Runge-Kutta Time-Stepping Schemes”, AIAA paper 81-1259, 1981.
    [18] Kuo, F.A.., Smith, M.R.., Hsieh, C.W., Chou, C.Y., Wu, J.S., “GPU acceleration for general conservation equations and its application to several engineering problems”, J. Compt. & Fluids, Vol. 45, pp. 147-154, 2011.
    [19] Lawson, G., Sosonkina, M., Yuzhong Shen, “Energy Evaluation for Applications with Different Thread Affinities on the Intel Xeon Phi”, 5th Workshop on Applications for Multi-Core Architectures (WAMCA 2014), Paris, Oct., 2014.
    [20] Lawson, G., Sosonkina, M., Yuzhong Shen, “Performance and Energy Evaluation of CoMD on Intel Xeon Phi Co-processors”, Hardware-Software Co-Design for High Performance Computing (Co-HPC), 2014, pp. 49 - 54.
    [21] Lim, C. W., Smith M.R., Jermy, M. C., Wu, C.-S., & Krumdieck, S. P., “The direction decoupled Quiet Direct Simulation method for rapid simulation of axisymmetric inviscid unsteady flow in Pulsed Pressure Chemical Vapour Deposition”, Computers & Fluids, 2013.
    [22] Lin, YJ, Smith, M.R., Kuo, F.A., Cave, H.M., Huang, J.C., Wu, J.S., “A true directional reconstruction to the quiet direct simulation method”, Computer Physics Communications, 184(11):2378–90, 2013.
    [23] Liu, J.Y., Smith M.R., Kuo, F.A., Wu, J.S., “Hybrid OpenMP/AVX acceleration of a Split HLL Finite Volume Method for the Shallow Water and Euler Equations”, Computers & Fluids, Vol. 110, p181-188. 8p. 2015
    [24] Macrossan, M.N., “The equilibrium flux method for the calculation of flows with nonequilibrium chemical reactions”, Journal of Computational Physics 80: 204-231, 1989.
    [25] Microprocessor Logic - HowStuffWorks. (n.d.). Retrieved May 3, 2015, from http://computer.howstuffworks.com/microprocessor2.htm.
    [26] Peery, K. M. & Imlay, S. T., “Blunt body flow simulations”, AIAA Paper 88-2924, 1988.
    [27] Pulliam, T.H., “Efficient Solution Methods for The Navier-Stokes Equations”, von Kármán Institute, Rhode-St-Genese, Belgium, 1985.
    [28] Pullin, D.I., “Direct Simulation Methods for Compressible Ideal Gas Flow”, Journal of Computational Physics 34: 231-244, 1980.
    [29] “Quick-Reference Guide to Optimization with Intel Compilers version 12”, Intel Corporation, 2010.
    [30] Reinders, J., “An Overview of Programming for Intel Xeon processors and Intel Xeon Phi coprocessors”, Intel Corporation, 2012.
    [31] Roger Espasa, Mateo Valero, “Exploiting Instruction- and Data-Level Parallelism”, IEEE Micro, pp. 20-27, Sep./Oct. 1997.
    [32] Sedov, L.I., “Similarity and Dimensional Methods in Mechanics”, Academic Press, New York, 1959.
    [33] Shaw, C.T., “Using Computational Fluid Dynamics”, Prentice Hall, New Jersey, USA. 1992.
    [34] Smith, M.R., Cave H, Wu JS, Jermy MC, Chen YS., “An improved quiet direct simulation method for Eulerian fluids using a second-order scheme”, J Comp. Phys. 228:2213–24, 2009.
    [35] Smith, M.R., Kuo, F.A., Hsieh, C.W., J.-P. Yu, Wu, J.S., and A. Ferguson, “Rapid Optimization of blast-wave mitigation strategies using Quiet Direct Simulation and Genetic Algorithm”, Computer Physics Communications, 181: pp. 1025–1036, 2010.
    [36] Smith, M.R., Liu, J.Y., Kuo, F.A., Wu, J.S., “Hybrid OpenMP/AVX acceleration of a higher order quiet direct simulation method for the Euler equations”, Procedia Engineering, Volume 61, Pages 152-157, 2013.
    [37] Sod, G.A., “A Survey of Several Finite Difference Methods for Systems of Nonlinear Hyperbolic Conservation Laws”, J. Comp. Phys., Vol. 27, pp. 1-31, 1978.
    [38] Toro, E.F., “Riemann Solvers and Numerical Methods for Fluid Dynamics”, Springer, Berlin, 2009.
    [39] van den Berg, A.C., Weerheijm, J., “Blast phenomena in urban tunnel systems”, Journal of Loss Prevention in the Process Industries, 19, 2006.
    [40] Versteeg, H. K., Malalasekera, W., “An Introduction to Computational Fluid Dynamics - The Finite Volume Method”, Pearson, Prentice Hall, 2007.
    [41] W.H. Beyer, “CRC Standard Mathematical Tables”, CRC Press, Boca Raton (FL), 1987.
    [42] Woodward, P. and Collela, P., “The Numerical simulation of two dimensional fluid flow with strong shocks”, Journal of Computational Physics, 54:115-173, 1984.

    無法下載圖示 校內：立即公開
    校外：不公開
    電子論文尚未授權公開，紙本請查館藏目錄
    QR CODE