基于面向对象的分布式遥感数据存储模型研究

Research on Remote Sensing Data Distributed Storage Model Based on Object-Oriented

现有的遥感数据存储模型大都针对遥感影像产品数据,且在分布式环境中不能达到较好的负载均衡,为了能够管理及高效存取海量多源异构的遥感数据,满足日益增长的高并发数据共享服务需求,提出了一种支持分布式虚拟化技术的遥感数据存储组织模型.模型利用面向对象思想对遥感数据进行分类及标准化处理,结合金字塔与全球剖分格网模型,建立分布式虚拟化环境下的遥感数据存储组织架构,利用一致性哈希算法思想解决数据在分布式环境下均衡性存储问题,同时通过虚拟化服务器动态分配策略优化了系统整体性能.实验表明新提出的模型比传统模型及同类模型更具优势.该模型已经应用在国家高分辨率对地观测系统河南数据与应用中心数据分发服务系统,验证了模型的可靠性、稳定性和适用性.

Current remote sensing data distributed storage models are not suitable to manage and storage mass multisource remote sensing data,due to the lack of better load balancing.In order to solve this problem,improve the speed of data accessing and meet the requirment of high concurrency data sharing service,a remote sensing data storage model based on object-oriented and distributed virtualization technology（RSDO）is proposed.This model used object-oriented to do data classification and standardization,and combined with Pyramid Model and Global Subdivision Grid Model to do data storage,and then the remote sensing data storage structure was constructed under the distributed environment with the improved consistent hash algorithm.It solves balance storage problem,and optimizes system integral performance with the virtual server dynamic allocation policy.This model has been used in Henan Data and Application Center of Earth Observation System,which reflects its reliability,stability and practicability.At last,the performance test and load balance experiment indicate that the proposed model has advantages over other traditional models.