李林,顾进锋,宋安捷,郑海宁,曹津,朱德海.基于GPU的生态环境遥感评价模型并行化研究[J].农业机械学报,2017,48(5):135-141.
LI Lin,GU Jinfeng,SONG Anjie,ZHENG Haining,CAO Jin,ZHU Dehai.Parallelization on Model of Ecological Environment Remote Sensing Evaluation Based on GPU[J].Transactions of the Chinese Society for Agricultural Machinery,2017,48(5):135-141.
摘要点击次数: 2389
全文下载次数: 1132
基于GPU的生态环境遥感评价模型并行化研究   [下载全文]
Parallelization on Model of Ecological Environment Remote Sensing Evaluation Based on GPU   [Download Pdf][in English]
投稿时间:2016-08-31  
DOI:10.6041/j.issn.1000-1298.2017.05.016
中文关键词:  生态环境评价  遥感  并行化处理  GPU  CUDA
基金项目:国家自然科学基金项目(31471762)
作者单位
李林 中国农业大学 
顾进锋 中国农业大学 
宋安捷 谢菲尔德大学 
郑海宁 中国农业大学 
曹津 中国农业大学 
朱德海 中国农业大学 
中文摘要:通过基于GPU的生态环境遥感评价模型并行化研究,在深入分析CPU+GPU异构通用计算平台数据传输瓶颈的基础上,设计了数据分片、异步传输的GPU图像处理框架。在此基础上,将碳固定量、草地退化指数和生态环境指数3个计算模型基于CUDA进行并行化实现,并通过实验验证了该技术方法的有效性,随着数据规模的变大,碳固定量计算模型的加速比达到了8.04倍,草地退化指数计算模型的加速比达到了12.21倍,生态环境指数计算模型的加速比达到了7.45倍。
LI Lin  GU Jinfeng  SONG Anjie  ZHENG Haining  CAO Jin  ZHU Dehai
China Agricultural University,China Agricultural University,University of Sheffield,China Agricultural University,China Agricultural University and China Agricultural University
Key Words:ecological environment evaluation  remote sensing  parallel processing  GPU  CUDA
Abstract:In order to solve the problem of the model’s slowly processing speed of ecological environment remote sensing evaluation currently, a framework about GPU image processing was designed with data partitioning and scheduling asynchronous transmission which was based on the in-depth analysis about the data transmission bottleneck of heterogeneous CPU+GPU general computing platform. It included the carbon fixed quantity and grassland degradation index, whose intrinsic parallelism met the GPU computing features. For the above models, it was put forward based on CUDA parallel implementation. The core link of indexes for evaluation of ecological environment of remote sensing data standardization and weighted fusion of CUDA parallel module were implemented. Finally, the effectiveness of technical methods was verified through experiments, as the scale of data became larger, the parallel execution speed of three business models became faster, the speedup ratio of the fixed amount of carbon achieved a 8.04 times execution rate lift;the speedup ratio of the index of grassland degradation achieved a 12.21 times execution rate lift;and the speedup ratio of the index of ecological environment achieved a 7.45 times execution rate lift. At the same time, the speedup ratio was decreased as the number of input data files increased, equipment between I/O was still the main factor which restricted the running efficiency of the algorithm.

Transactions of the Chinese Society for Agriculture Machinery (CSAM), in charged of China Association for Science and Technology (CAST), sponsored by CSAM and Chinese Academy of Agricultural Mechanization Science(CAAMS), started publication in 1957. It is the earliest interdisciplinary journal in Chinese which combines agricultural and engineering. It always closely grasps the development direction of agriculture engineering disciplines and the published papers represent the highest academic level of agriculture engineering in China. Currently, nearly 8,000 papers have been already published. There are around 3,000 papers contributed to the journal each year, but only around 600 of them will be accepted. Transactions of CSAM focuses on a wide range of agricultural machinery, irrigation, electronics, robotics, agro-products engineering, biological energy, agricultural structures and environment and more. Subjects in Transactions of the CSAM have been embodied by many internationally well-known index systems, such as: EI Compendex, CA, CSA, etc.

   下载PDF阅读器