面向过程的海洋时空数据分布式存储与并行检索  

Process-Oriented Distributed Storage and Retrieval of Ocean Spatiotemporal Data

作　　者：谭凯中 秦勃[1] 何亚文[2] Tan Kaizhong;Qin Bo;He Yawen(College of Information Science and Engineering, Ocean University of China, Qingdao 266100, China;College of Oceanography and Space Informatics, China University of Petroleum, Qingdao 266580, China)

机构地区：[1]中国海洋大学信息科学与工程学院,山东青岛266100 [2]中国石油大学(华东)海洋与空间信息学院,山东青岛266580

出　　处：《中国海洋大学学报（自然科学版）》2021年第11期94-101,134,共9页Periodical of Ocean University of China

基　　金：国家自然科学基金项目(41976184)资助。

摘　　要：海洋现象特征数据是一类带有过程性质的时空数据,具有多样性、多态性、海量性和时变性等特点,广泛服务于海洋灾害预报、船舶航行规划、台风路径追踪、海洋与大气环流等领域。本文根据海洋现象特征,建立海洋现象过程的对象表达和数据组织模型,设计三维空间海洋时空数据分片策略和R树空间数据索引方法,以提高海洋现象过程时空数据的查询效率。实验通过构建海洋现象过程时空数据分布式存储与并行检索系统测试算法的执行效率。实验结果表明,面向过程的海洋时空数据分布式并行检索算法能够实现数据的高效检索,验证了海洋现象过程对象数据组织模型的合理性,为研究海洋现象过程多要素的关联机制提供了有效的方法和理论基础。Ocean phenomenon feature data is a kind of spatiotemporal data with process nature,which has the characteristics of diversity,polymorphism,mass and time variability.It is widely used in Marine disaster prediction,ship navigation planning,typhoon path tracking,atmospheric circulation.According to the features of ocean phenomena,design object expression and data model of ocean phenomena process,uses 3D ocean spatiotemporal data fragmenting strategy and R-tree spatial data index technology to improve query efficiency of ocean phenomena process spatial-temporal data.The experiment tested the efficiency of the algorithm by building distributed retrieval and storage system of ocean phenomena process spatial-temporal data.The experimental results show that the process-oriented distributed parallel retrieval algorithm of ocean spatiotemporal data has efficient data retrieval performance,validate the rationality of the ocean phenomenal process objects data model,and provide effective theoretical basis for studying the multiple elements association mechanism of ocean phenomenal process.

关 键 词：面向过程 海洋时空数据 分布式检索 分布式存储 R树空间数据索引 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]