摘要
空间信息技术和遥感遥测等技术的飞速发展,产生了海量的遥感、地灾等行业空间信息数据。如何对海量空间数据进行合理的分级存储,以满足大数据时代下空间信息、地理信息等行业应用,这已成为日益紧迫的问题。海量空间数据分级存储作为一种全新的存储模式,为解决该问题提出了新的思路。结合海量空间数据的特点和日常数据应用的规律,提出了基于访问热度和聚类关联的海量空间数据分级存储模型,该模型主要包括热点数据分级、关联数据分级、数据的迁移三部分。最后通过嫦娥2号遥感数据模拟访问试验,优化了数据升级阀值,证明了分级存储模型用于海量空间数据的可行性。
With the rapid development of space information technology and remote sensing technology,vast amounts of spatial information data like remote sensing and geological disasterwere produce.How reasonable tier stored massive spatial data to make meet the needs of applications like spatial information and geographic information is becoming an increasingly urgent problem.Hierarchical storage massive spatial data as a new model for solving the problem put forward new ideas.Combined with the characteristics of massive spatial data and rule of daily data application,put forward the data presented hierarchical data storage model and the associated heat-based access massive spatial clustering,the model includes hot data classification,association data classification,data migration in three parts.Finally,simulation access testing on Change 2remote sensing data optimized the threshold of data upgrade;it proved the feasibility of the tiered storage model for spatial data.
出处
《物探化探计算技术》
CAS
CSCD
2015年第6期783-789,共7页
Computing Techniques For Geophysical and Geochemical Exploration
基金
国家自然科学基金项目(61071121)
成都市经信委科技专项项目(201102153)
关键词
空间数据
遥感遥测
分级存储
密度聚类算法
数据关联
spatial data
remote Sensing
tiered storage
density clustering algorithm
data association
