
浏览全部资源
扫码关注微信
1.军事科学院国防科技创新研究院,北京,100071
2.信息工程大学地理空间信息学院,河南 郑州,450001
3.96941部队,北京,100080
Received:06 September 2023,
Published:05 March 2025
移动端阅览
黄炎,李姗姗,吕明昊,等.利用CPU和GPU混合并行方法快速构建海洋扰动重力梯度基准图[J].武汉大学学报(信息科学版),2025,50(3):515-527.
HUANG Yan,LI Shanshan ,LV Minghao,et al.Rapid Construction of Marine Disturbing Gravity Gradient Reference Map by CPU and GPU Hybrid Parallel Method[J].Geomatics and Information Science of Wuhan University,2025,50(3):515-527.
黄炎,李姗姗,吕明昊,等.利用CPU和GPU混合并行方法快速构建海洋扰动重力梯度基准图[J].武汉大学学报(信息科学版),2025,50(3):515-527. DOI:10.13203/j.whugis20220142
HUANG Yan,LI Shanshan ,LV Minghao,et al.Rapid Construction of Marine Disturbing Gravity Gradient Reference Map by CPU and GPU Hybrid Parallel Method[J].Geomatics and Information Science of Wuhan University,2025,50(3):515-527. DOI:10.13203/j.whugis20220142
高精度、高分辨率的海洋扰动重力梯度基准图是将水下扰动重力梯度辅助惯性导航付诸于实践的关键技术之一,可依据边值问题理论,基于移去-恢复技术计算求得。基于传统串行算法存在计算效率低、耗时长等问题,为满足大范围乃至全球海洋扰动重力梯度基准图的快速构建需求,利用中央处理器(central processing unit
CPU)与图形处理器(graph processing unit
GPU)混合并行(CPU+GPU)编程平台,设计了一种高效的混合并行计算方案。首先,利用数组收缩膨胀方法,有效解决了CPU进行勒让德函数递推计算过程中内存读写冲突问题;然后,引入Hilbert空间填充曲线,将二维格网重力异常数据降维成为具有高度聚簇性的一维数组,通过其在GPU显存中的快速索引,实现了Stokes积分的高效计算。在CPU型号为Intel Xeon(R) Gold 6130、GPU型号为Tesla V100的计算机上进行实验,利用该混合并行方案计算6°×6°范围、1′分辨率的全张量海洋扰动重力梯度基准图仅需15.84 s,而传统串行方法需要35 min以上;计算2°×2°范围、30″分辨率基准图仅需22.7 s,串行方法则需要52 min;计算2°×2°范围、10″分辨率基准图需要26.7 min,串行方法则需要两天以上。在保证并行计算绝对误差小于等于1×10
-6
E、
<math id="M1"><msub><mrow><mi>T</mi></mrow><mrow><mi>x</mi><mi>x</mi></mrow></msub><mo>+</mo><msub><mrow><mi>T</mi></mrow><mrow><mi>y</mi><mi>y</mi></mrow></msub><mo>+</mo><msub><mrow><mi>T</mi></mrow><mrow><mi>z</mi><mi>z</mi></mrow></msub></math>
https://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=76268667&type=
3.30200005
https://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=76268680&type=
20.40466690
绝对值的最大值小于0.1 E、均方根误差小于0.01 E的前提下,实现了全球全张量海洋扰动重力梯度基准图模型的快速构建。
Objectives
2
High precision and high resolution marine disturbing gravity gradient reference map is one of the key technologies to put underwater disturbing gravity gradient assisted inertial navigation into practice. At present
gravity anomaly is mainly measured by satellite and calculated based on the boundary value problem theory and the removal recovery technique.
Methods
2
The traditional serial algorithm has some problems such as low efficiency and long time consuming. In order to meet the needs of rapid construction of marine disturbing gravity gradient reference maps on a large scale and even around the world
this paper designs an efficient hybrid parallel computing scheme using central processing unit (CPU)+graph processing unit (GPU) hybrid parallel programming platform. First
the shrink and expansion method of array is proposed to solve the problem of memory conflict between read and write in Legendre recursive computation. Then Hilbert space filling curve is used to reduce the dimension of two-dimensional grid gravity anomaly data into a one-dimensional array with high clustering. The efficient calculation of Stokes integral is realized by the fast index in GPU video memory.
Results
2
The experiments are carried out on computers with Intel Xeon (R) Gold 6130 CPU and Tesla V100 GPU. The hybrid parallel scheme is used to calculate 6°×6° range and 1' resolution full tensor marine disturbing gravity gradient reference map and only takes 15.84 seconds
while the traditional serial method takes more than 35 minutes. To calculate 2°×2° range and 30'' resolution map
the proposed method only takes 22.7 seconds
while the serial method takes 52 minutes. To calculate 2°×2° range and 10'' resolution map
the proposed method only takes 26.7 minutes
while the serial method takes more than two days.
Conclusions
2
On the premise of ensuring that the absolute error of parallel computing is less than or equal to 1×10
-6
E
the maximum absolute value of
<math id="M2"><msub><mrow><mi>T</mi></mrow><mrow><mi>x</mi><mi>x</mi></mrow></msub><mo>+</mo><msub><mrow><mi>T</mi></mrow><mrow><mi>y</mi><mi>y</mi></mrow></msub><mo>+</mo><msub><mrow><mi>T</mi></mrow><mrow><mi>z</mi><mi>z</mi></mrow></msub></math>
https://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=76268699&type=
3.80999994
https://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=76268716&type=
23.62199974
is less than 0.1 E and the root mean square error is less than 0.01 E
the rapid construction of the global total tensor marine disturbing gravity gradient reference map model is realized.
赵伟 . 实时重力梯度补偿的惯性导航方法研究 [D]. 武汉 : 武汉大学 , 2014 .
ZHAO Wei . Research on Inertial Navigation Method with Real-Time Gravity Gradient Compensation [D]. Wuhan : Wuhan University , 2014 .
LI Z W , ZHENG W , WU F . Improving the Reliabi-lity of Underwater Gravity Matching Navigation Based on a Priori Recursive Iterative Least Squares Mismatching Correction Method [J]. IEEE Access , 2020 , 8 : 8648 - 8657 .
YOO Y M , LEE W H , LEE S M , et al . Improvement of TERCOM Aided Inertial Navigation System by Velocity Correction [C]//IEEE/ION Position, Location and Navigation Symposium. Myrtle Beach , USA , 2012 .
王博 , 付梦印 , 李晓平 , 等 . 水下重力匹配定位算法综述 [J]. 导航与控制 , 2020 , 19 ( S1 ): 170 - 178 .
WANG Bo , FU Mengyin , LI Xiaoping , et al . Overview of Underwater Gravity Matching Positioning Algorithms [J]. Navigation and Control , 2020 , 19 ( S1 ): 170 - 178 .
赵伟 , 王正涛 , 马强 . 重力梯度数据模拟及其在惯性导航中的应用 [J]. 大地测量与地球动力学 , 2014 , 34 ( 5 ): 134 - 138 .
ZHAO Wei , WANG Zhengtao , MA Qiang . Simulation of Gravity Gradient Data and Its Applications on Inertial Navigation [J]. Journal of Geodesy and Geodynamics , 2014 , 34 ( 5 ): 134 - 138 .
黄炎 . 基于并行计算的重力场若干问题方法研究 [D]. 郑州 : 信息工程大学 , 2020 .
HUANG Yan . Research on Several Problems of Gravity Field Based on Parallel Computing [D]. Zhengzhou : Information Engineering University , 2020 .
黄佳喜 . 扰动重力场快速(并行)计算方法研究 [D]. 郑州 : 信息工程大学 , 2017 .
HUANG Jiaxi . Study on Fast (Parallel) Calculation Method of Disturbed Gravity Field [D]. Zhengzhou : Information Engineering University , 2017 .
都志辉 . 高性能计算并行编程技术: MPI并行程序设计 [M]. 北京 : 清华大学出版社 , 2001 .
DU Zhihui . Parallel Programming Technology for High Performance Computing: MPI Parallel Programming [M]. Beijing : Tsinghua University Press , 2001 .
周雪 , 于平 , 翁爱华 , 等 . 基于MPI和OpenMP的重力及重力梯度数据并行正演算法研究 [J]. 世界地质 , 2018 , 37 ( 3 ): 897 - 904 .
ZHOU Xue , YU Ping , WENG Aihua , et al . Parallel Forward Modelling Algorithm with Gravity and Gravity Gradient Data Based on MPI and OpenMP [J]. Global Geology , 2018 , 37 ( 3 ): 897 - 904 .
张晗 , 钱育蓉 , 侯海耀 . CUDA下地质图像边缘检测算法并行优化 [J]. 计算机工程与设计 , 2019 , 40 ( 3 ): 691 - 698 .
ZHANG Han , QIAN Yurong , HOU Haiyao . Parallel Optimization of Edge Detection Algorithm for Geological Images Under CUDA [J]. Computer Engineering and Design , 2019 , 40 ( 3 ): 691 - 698 .
NVIDIA Technical Report . Nvidia® Tesla® p100–the Most Advanced Data Center Accelerator Ever Built [R]. Santa Clara, USA , 2017 .
李午阳 , 张健 , 林巍 . 基于CUDA的GPU并行优化重力三维反演 [J]. 物探与化探 , 2016 , 40 ( 1 ): 179 - 184 .
LI Wuyang , ZHANG Jian , LIN Wei . Regularized Inversion of 3D Gravity Data: A New GPU Parallelized Method Based on CUDA [J]. Geophysical and Geochemical Exploration , 2016 , 40 ( 1 ): 179 - 184 .
SHARMA G , AGARWALA A , BHATTACHARYA B . A Fast Parallel Gauss Jordan Algorithm for Matrix Inversion Using CUDA [J]. Computers & Structures , 2013 , 128 : 31 - 37 .
张舒 , 褚艳利 . GPU高性能运算之CUDA [M]. 北京 : 中国水利水电出版社 , 2009 .
ZHANG Shu , CHU Yanli . CUDA for GPU High Performance Computing [M]. Beijing : China Water & Power Press , 2009 .
赵炳杰 , 杨光亮 , 黎哲君 . 移去-恢复法校准重力异常场及其应用 [J]. 大地测量与地球动力学 , 2019 , 39 ( 4 ): 361 - 365 .
ZHAO Bingjie , YANG Guangliang , LI Zhejun . Study on the Remove-Restore Theory of Refining Gravity Anomaly and Its Application [J]. Journal of Geodesy and Geodynamics , 2019 , 39 ( 4 ): 361 - 365 .
陈涛 , 张贵宾 . 利用重力异常计算重力梯度的等效源技术 [J]. 地球物理学进展 , 2019 , 34 ( 4 ): 1398 - 1410 .
CHEN Tao , ZHANG Guibin . Deriving the Full Gravitational Gradient Tensor from Gravity Anomaly: An Equivalent Source Technique [J]. Progress in Geophysics , 2019 , 34 ( 4 ): 1398 - 1410 .
汪凤林 , 蔡体菁 , 王东霞 . 惯性/重力匹配组合导航系统可观测性研究 [J]. 安徽大学学报(自然科学版) , 2008 , 32 ( 6 ): 27 - 31 .
WANG Fenglin , CAI Tijing , WANG Dongxia . Inertial/Gravity Matching Integrated Navigation System [J]. Journal of Anhui University (Natural Sciences) , 2008 , 32 ( 6 ): 27 - 31 .
杨靖宇 , 张永生 , 李正国 , 等 . 遥感影像正射纠正的GPU-CPU协同处理研究 [J]. 武汉大学学报(信息科学版) , 2011 , 36 ( 9 ): 1043 - 1046 .
YANG Jingyu , ZHANG Yongsheng , LI Zhengguo , et al . GPU-CPU Cooperate Processing of RS Image Ortho-Rectification [J]. Geomatics and Information Science of Wuhan University , 2011 , 36 ( 9 ): 1043 - 1046 .
陈召曦 , 孟小红 , 郭良辉 , 等 . 基于GPU并行的重力、重力梯度三维正演快速计算及反演策略 [J]. 地球物理学报 , 2012 , 55 ( 12 ): 4069 - 4077 .
CHEN Zhaoxi , MENG Xiaohong , GUO Lianghui , et al . Three-Dimensional Fast Forward Modeling and the Inversion Strategy for Large Scale Gravity and Gravimetry Data Based on GPU [J]. Chinese Journal of Geophysics , 2012 , 55 ( 12 ): 4069 - 4077 .
刘文志 . 并行算法设计与性能优化 [M]. 北京 : 机械工业出版社 , 2015 .
LIU Wenzhi . Parallel Computing and Performance Optimization [M]. Beijing : China Machine Press , 2015 .
许大欣 . 利用重力异常匹配技术实现潜艇导航 [J]. 地球物理学报 , 2005 , 48 ( 4 ): 812 - 816 .
XU Daxin . Using Gravity Anomaly Matching Techniques to Implement Submarine Navigation [J]. Chinese Journal of Geophysics , 2005 , 48 ( 4 ): 812 - 816 .
曹雪峰 , 万刚 , 张宗佩 . Hilbert曲线层级演进关系 [J]. 测绘学报 , 2016 , 45 ( S1 ): 77 - 84 .
CAO Xuefeng , WAN Gang , ZHANG Zongpei . Hilbert Curve Hierarchical Evolution Relution [J]. Acta Geodaetica et Cartographica Sinica , 2016 , 45 ( S1 ): 77 - 84 .
刘辉 , 冷伟 , 崔涛 . 高维Hilbert曲线的编码与解码算法设计 [J]. 数值计算与计算机应用 , 2015 , 36 ( 1 ): 42 - 58 .
LIU Hui , LENG Wei , CUI Tao . Development of Encoding and Decoding Algorithms for High Dimensional Hilbert Curves [J]. Journal on Numerical Methods and Computer Applications , 2015 , 36 ( 1 ): 42 - 58 .
贾连印 , 陈明鲜 , 李孟娟 , 等 . 基于状态视图的高效Hilbert编码和解码算法 [J]. 电子与信息学报 , 2020 , 42 ( 6 ): 1494 - 1501 .
JIA Lianyin , CHEN Mingxian , LI Mengjuan , et al . State View Based Efficient Hilbert Encoding and Decoding Algorithms [J]. Journal of Electronics & Information Technology , 2020 , 42 ( 6 ): 1494 - 1501 .
吴宇豪 , 曹雪峰 . 虚拟战场环境时空数据的Hilbert码索引方法 [J]. 武汉大学学报(信息科学版) , 2020 , 45 ( 9 ): 1403 - 1411 .
WU Yuhao , CAO Xuefeng . Hilbert Code Index Method for Spatiotemporal Data of Virtual Battlefield Environment [J]. Geomatics and Information Science of Wuhan University , 2020 , 45 ( 9 ): 1403 - 1411 .
孙立伟 , 袁昱纬 , 周俊芳 . 一种采用Hilbert空间排列码的场景数据调度策略 [J]. 无线电工程 , 2019 , 49 ( 9 ): 775 - 778 .
SUN Liwei , YUAN Yuwei , ZHOU Junfang . A Scene Data Scheduling Strategy Using Hilbert Space Arrangement Codes [J]. Radio Engineering , 2019 , 49 ( 9 ): 775 - 778 .
ODERA P A , FUKUDA Y , KUROISHI Y . A High-Resolution Gravimetric Geoid Model for Japan from EGM2008 and Local Gravity Data [J]. Earth, Planets and Space , 2012 , 64 ( 5 ): 361 - 368 .
范宏涛 , 郭春喜 , 王小华 , 等 . 超高阶重力场模型EIGEN-6C2适应性分析 [J]. 测绘科学 , 2015 , 40 ( 9 ): 18 - 22 .
FAN Hongtao , GUO Chunxi , WANG Xiaohua , et al . Adaptability Analysis on Ultra-High-Gravity Model EIGEN-6C2 in Chinese Coastal Zone [J]. Science of Surveying and Mapping , 2015 , 40 ( 9 ): 18 - 22 .
0
Views
0
下载量
0
CSCD
Publicity Resources
Related Articles
Related Author
Related Institution
京公网安备11010802024621