Communication Reduced Multi-time-step Algorithm for Real-time Wind Simulation on GPU-based Supercomputers

被引：9
|
作者：
Onodera, Naoyuki [1 ]
Idomura, Yasuhiro [1 ]
Ali, Yussuf [1 ]
Shimokawabe, Takashi [2 ]
机构：
[1] Japan Atom Energy Agcy, Kashiwa, Chiba, Japan
[2] Univ Tokyo, Bunkyo Ku, Tokyo, Japan
来源：
SCALA 2018: PROCEEDINGS OF 2018 IEEE/ACM 9TH WORKSHOP ON LATEST ADVANCES IN SCALABLE ALGORITHMS FOR LARGE-SCALE SYSTEMS (SCALA) | 2018年
基金：
日本学术振兴会;
关键词：
Locally Mesh Refinement; Communication Reduced Multi-time-step Algorithm; GPU; Wind Simulation; LATTICE BOLTZMANN METHOD; ADAPTIVE MESH REFINEMENT; GRID-REFINEMENT; TURBULENT FLOWS;
D O I：
10.1109/ScalA.2018.00005
中图分类号：
TP301 [理论、方法];
学科分类号：
081202 ;
摘要：
We develop a communication reduced multi-timestep (CRMT) algorithm for a Lattice Boltzmann method (LBM) based on a block-structured adaptive mesh refinement (AMR). This algorithm is based on the temporal blocking method, and can improve computational efficiency by replacing a communication bottleneck with additional computation. The proposed method is implemented on an extreme scale airflow simulation code CityLBM, and its impact on the scalability is tested on GPU based supercomputers, TSUBAME and Reedbush. Thanks to the CRMT algorithm, the communication cost is reduced by similar to 64%, and weak and strong scalings are improved up to similar to 200 GPUs. The obtained performance indicates that real time airflow simulations for about 2km square area with the wind speed of similar to 5m/s is feasible using 1m resolution.
引用
收藏
页码：9 / 16
页数：8
相关论文
共 50 条
  • [1] Real-time GPU-based simulation of dynamic terrain
    Aquilio, Anthony S.
    Brooks, Jeremy C.
    Zhu, Ying
    Owen, G. Scott
    [J]. ADVANCES IN VISUAL COMPUTING, PT 1, 2006, 4291 : 891 - +
  • [2] GPU-based Morphing Terrain Real-time Shadow Algorithm
    Wang Xiaojun
    Zhai Haitao
    Xia Jie
    Cui Zhiming
    [J]. 2009 INTERNATIONAL SYMPOSIUM ON WEB INFORMATION SYSTEMS AND APPLICATIONS, PROCEEDINGS, 2009, : 198 - 201
  • [3] A GPU-based Method for Real-time Simulation of Eastern Painting
    Lu, The Kiet
    Huang, Zhiyong
    [J]. GRAPHITE 2007: 5TH INTERNATIONAL CONFERENCE ON COMPUTER GRAPHICS AND INTERACTIVE TECHNIQUES IN AUSTRALASIA AND SOUTHERN ASIA, PROCEEDINGS, 2007, : 111 - 118
  • [4] GPU-based real-time crowd rendering
    Zhang, Yong
    Yin, Bao-Cai
    Kong, De-Hui
    Yang, Guang-Wei
    [J]. Beijing Gongye Daxue Xuebao / Journal of Beijing University of Technology, 2009, 35 (10): : 1434 - 1440
  • [5] GPU-based Real-time Face Detector
    Jeong, Jae-chan
    Shin, Ho-chul
    Cho, Jae-il
    [J]. 2012 9TH INTERNATIONAL CONFERENCE ON UBIQUITOUS ROBOTS AND AMBIENT INTELLIGENCE (URAL), 2012, : 173 - 175
  • [6] GPU-based real-time simulation and rendering of unbounded ocean surface
    Yang, XD
    Pi, XX
    Zeng, L
    Li, SK
    [J]. NINTH INTERNATIONAL CONFERENCE ON COMPUTER AIDED DESIGN AND COMPUTER GRAPHICS, PROCEEDINGS, 2005, : 428 - 433
  • [7] A GPU-based method for approximate real-time fluid flow simulation
    University of Science and Technology AGH, Institute of Computer Science, al. Mickiewicza 30, 30-059 Kraków, Poland
    [J]. Mach. Graph. Vis., 2008, 3 (267-278):
  • [8] A GPU-Based Multi-agent System for Real-Time Simulations
    Vigueras, Guillermo
    Orduna, Juan M.
    Lozano, Miguel
    [J]. ADVANCES IN PRACTICAL APPLICATIONS OF AGENTS AND MULTIAGENT SYSTEMS, 2010, 70 : 15 - 24
  • [9] GPU-based real-time deformation with normal reconstruction
    Che, Yinghui
    Wang, Jing
    Liang, Xiaohui
    [J]. TECHNOLOGIES FOR E-LEARNING AND DIGITAL ENTERTAINMENT, PROCEEDINGS, 2007, 4469 : 667 - +
  • [10] GPU-based real-time RGBD data filtering
    Abdenour Amamra
    Nabil Aouf
    [J]. Journal of Real-Time Image Processing, 2018, 14 : 323 - 340
12345