Research On The Storage Management Strategy Of Remote Sensing Data Base On MongoDB

As akind of earth observation target information carrier, remote sensing data has a large coverage area, information-rich content, access speed and other characteristics,and was widely used in forestry, geology, marine, agriculture, environmental protection, meteorology, mapping and military reconnaissance,etc. In recent years, with the rapid development of remote sensing technology, remote sensing data rendered progressively multi-source, multi-scale, multi-temporal, global coverage and high resolution features,formed a GB, TB and PB of development trend.On the other hand, in order to realize the real-time display and quick service of the massive remote sensing image information,thatmakes more use of multi-resolution image pyramid data block technology in the remote sensing applications, which has led to the amount of data of remote sensing data increases exponentially. How to be more efficient and orderly organization, storage and management of these massive remote sensing data has become a challenge for us.With the rapid development of computer and information technology,, the field of remote sensing continues to progress in. In recent years, massive remote sensing data storage management has been developed to a distributed cluster.By analyzing the advantages and disadvantages of popular domestic and foreign remote sensing mass data storage management system,for the requirement of the tile storage management module of China Remote Sensing Satellite Application Service System- integrated database operation and management subsystem,this paper has done the following research:(1) Study and design data storage management strategy by using MongoDB clustered to storage tile metadata anddistributed file system to storage the tile data file data; Based on tile GeoHash encode and using the MongoDB auto-sharding to let adjacent location data's metadata store in the same data block; In terms of meta data's retrieval, we design a range query strategy based on GeoHash encoding, compared with retrieval strategy based on level,row and col, has the certain promotion in the retrieval efficiency.(2) In the distributed file system, adopting the continuous distribution strategy, the construction of the absolute path to realize the direct addressing of tile data, greatly simplifies the process of addressing, implements the tiles file data addressing fast and accurate.(3) This paper studied remote sensing data storage management strategyhas been initially run in China Remote Sensing Satellite Application Service System- integrated database operation and management subsystem,and through experiments and test has run a certain time, prove the feasibility of this research strategy.