云存储是解决动态增长的海量遥感数据产品存储管理难题的有效手段.针对云存储在遥感数据存储领域中存在的问题,提出了一种基于影像块组织的遥感数据分布式Key-Value存储模型,解决了分布式文件系统存储大规模影像块效率低下的问题,使遥感数据云存储具备了空间区域访问特性;结合开源分布式文件系统HDFS,实现了影像数据的分布式高效存储与空间区域检索.实验与分析表明,系统在多用户并发连接情况下可以维持较高的吞吐率,同时具备良好的可伸缩性和稳定性.
As the volume of remote sensing data products is massive and growing rapidly in recent years, cloud stor- age with unparalleled scalability and high throughput is considered as a most potential way to solve the challenges of the storage and management, and reduce the investment for storage equipment. In this paper, the problems existing in the application of cloud storage for remote sensing data were brought up, and a distributed Key-Value storage module based on image blocks organization was put forward to settle them. Thus, the storage inefficiencies problem of distributed file system of cloud platforms for massive image blocks was solved, and the ability of spatial data ac- cess of the remote sensing data cloud storage was introduced. On the basis of HDFS, an open-source cloud-based distributed file system, the efficient distributed storage and retrieval of image data were implemented. The experi ment and analysis showed that the storage system maintains a high throughput under multiple concurrent connec- tions, having a good level of scalability and stability.