《计算流体力学相关解决方案PPT精选文档.ppt》由会员分享，可在线阅读，更多相关《计算流体力学相关解决方案PPT精选文档.ppt（16页珍藏版）》请在三一文库上搜索。
1、,TESLA,Computational Fluid Dynamics Module,Computational Fluid Dynamics,GPU Perf compared against Multi-core x86 CPU socket, features and may be a kernel to kernel perf comparison,GPU Value to Engineering Computational Fluid Dynamics,FluiDyna LBultra 20 x acceleration with 4 GPUs vs. 2 x 6 core CPUs
2、,CPU Intel Xeon X5670 2.93 GHz; GPU Tesla M2070,GPU READY APPLICATIONS Altair AcuSolve Autodesk Moldflow FluiDyna Culises for OpenFOAM FluiDyna LBultra Vratis SpeedIT for OpenFOAM Prometech Particleworks Sandia NL and ORNLS3D SD+ (SU-Jameson) FEFLO (GMU-Lohner) Turbostream,ANSYS CFD preliminary resu
3、lts of radiation heat transfer view-factor computations on GPUs vs. CPUs RHT on GPUs will release in 14.0 as beta Radiation HT Applications:,NOTE: Growing CPU time of view-factor computations inhibit proper inclusion of radiation HT effects,NOTE: GPU time remains low even as view-factor computations
4、 grow very large,ANSYS CFD 14.0 Offers First GPU Capability,Underhood cooling Cabin comfort HVAC Furnace simulations Solar loads on buildings Combustor in turbine Electronics passive cooling,Other ANSYS CFD Evaluations:,Models (e.g. disperse phase) Implicit equation solvers,OpenFOAM on GPUs ISVs Flu
5、iDyna and Vratis,Prometech and Particle-based CFD for Multi-GPUs,MPS-based method developed at the University of Tokyo Prof. Koshizuka Results shown for Particleworks 2.5 released in 2011 Performance is relative to 4 cores of Intel i7 CPU Contact Prometech for license details,http:/www.prometech.co.
6、jp,Turbostream CFD for Gas Turbine Engines,Turbostream Simulation Speed-up 19x,19x,www.turbostream-|,www.many-core.group.cam.ac.uk/ukgpucc2/talks/Brandvik.pdf|,Sources:,www.hpc.cam.ac.uk/services/darwin.html,University of Cambridge DARWIN Cluster,CUDA Center of Excellence Since 2008 GPU sub-cluster:
7、 Dell T5500 servers, 32 dual-socket CPUs Tesla S1070 GPUs, 4 GPUs per socket for total 128 GPUs,Sample Turbostream GPU Simulations,Typical Routine Simulation,Large-scale Simulation,19x Speedup,http:/www.turbostream-,Source:,Sample Turbostream GPU Simulations,Turbostream: CFD for Turbomachinery,Tokyo
8、 Institute of Technology AOKI Laboratory CFD Research on #5 of Top 500 TSUBAME 2.0 Simulations that scale to 4000 Fermi GPUs Presentation at Supercomputing 2010 Conference: “Large-scale CFD Applications on TSUBAME 2” Dr. Takayuki Aoki, Global Scientific Information and Computing Center (GSIC) of Tok
9、yo Institute of Technology (Tokyo Tech) ,GPU Highlights on CFD Applications From the Top 500,FEFLO: Porting of an Edge-Based CFD Solver to GPUs AIAA-2010-0523 Andrew Corrigan, Ph.D., Naval Research Lab; Rainald Lohner, Ph.D., GMU FAST3D: Using GPU on HPC Applications to Satisfy Low Power Computation
10、al Requirement AIAA-2010-0524 Gopal Patnaik, Ph.D., US Naval Research Lab OVERFLOW: Rotor Wake Modeling with a Coupled Eulerian and Vortex Particle Method AIAA-2010-0312 Chris Stone, Ph.D., Intelligent Light SOLAR: Unstructured CFD Solver on GPUs Jamil Appa, Ph.D., BAE Systems Advanced Technology Ce
11、ntre elsA: Recent Results with elsA on Many-Cores Michel Gazaix and Steve Champagneux, ONERA / Airbus France Turbostream: Turbostream: A CFD Solver for Many-Core Processors Tobias Brandvik, Ph.D. , Whittle Lab, University of Cambridge OVERFLOW: Acceleration of a CFD Code with a GPU Dennis Jespersen,
12、 NASA Ames Research Center,48th AIAA Aerospace Sciences Meeting | Jan 2010 | Orlando, FL, USA,CFD on Future Architectures | Oct 2009 | DLR Braunschweig, DE,Parallel CFD 2009 | May 2009 | NASA Ames, Moffett Field, CA, USA,Published CFD Developments on Tesla GPU,Total 110 technical papers: 32 or 30% i
13、ncluded GPU-developments, up from 12 papers in 2010 (Taipei, TW) and 4 papers in 2009 (NASA, US) Included an invited full-day workshop on CUDA and GPUs for CFD Applications attended by more than 100 delegates GPUs in talks from 6 of 7 plenary speakers: GPU-specific CFD: Aoki Tokyo Inst Tech, Lohner
14、GMU, Barber BU GPU evaluation: Chalot Dassault Aviation, Gonzlez Next Limit, Jgerskpper DLR,GPUs Highlights from ParCFD 2011,23rd ParCFD 2011 | 16 20 May 2011 | Barcelona, ES,GPU Application Jameson-developed CFD software SD+ for high order method aerodynamic simulations,GPU Benefit Use of 16 x Tesl
15、a M2070: 15 hrs vs. 202 hrs for 16 x Xeon X5670 Fast turnaround of complex LES simulations that would otherwise be impractical for CPU-only use,Stanford University Aerospace Computing Lab Prof. Antony Jameson,High-Order Aerodynamics Research on GPUs,15 hours on 16 x M2070s 202 hours ( one week) on 1
16、6 Xeon x5670 CPUs,Transitional flow over SD70053 airfoil, 21M DOF, Ma =.2, Re=60K, AoA=4, 4th order, 400K RK iters,GPU Application SJTU-developed CFD software NUS3D for aerodynamic simulations of wing shapes,GPU Benefit Use of Tesla C2070: 20 x 37x vs. single core Intel core i7 CPU Faster simulation
17、s for more wing design candidates vs. wind tunnel testing Expanding to multi-GPU and full aircraft,COMAC and SJTU Commercial Aircraft Corporation of China,COMAC Wing Candidate,ONERA M6 Wing CFD Simulation,Commercial Aircraft Wing Design on GPUs,GPU Application BAE-developed CFD software Veloxi for a
18、erodynamic simulation of aircraft,GPU Benefit Use of 2 x Tesla C2050: 15x vs. QC Intel i7 CPU Faster simulations enabled design exploration of full aerodynamic envelope,GPU Speed-up vs. Multi-core,BAE Systems Technology and Engineering Services,Defense Aerodynamic Applications on GPUs,GPU Applicatio
19、n NRL-developed CFD software JENRE for simulation of jet engine acoustics,GPU Benefit Use of Tesla M2070: 3x vs. Hex core Intel (Westmere) CPU More detailed mesh simulations possible for longer durations of jet engine transient conditions,U.S. DoD Naval Research Lab Lab for Computational Physics and
20、 Fluid Dynamics,Fighter Jet Engine Noise Reduction on GPUs,GPU Application EM Photonics-developed CFD software for unsteady aerodynamic simulations,GPU Benefit Use of Tesla C2070: 54x for CFD kernel vs. 8 core Intel i7 CPUs Fast turnaround of simulations enables more flight conditions and aircraft approach directions,NAVAIR and EM Photonics U.S. DoD Naval Air Weapons Center, Pax River MD,Aircraft Carrier Landing Effects using GPUs,54xfaster with 1 GPU,