The distributed storage strategy research of remote sensing image based on Mongo DB

In recent years, with the wisdom of the city promote and research in the field of geographic space unceasingly thoroughly, the mass spatial data storage management and multiple concurrent access has become the research hotspot in the field of geosciences. Traditional spatial data storage is based on relational database, modeling, index mechanism, cache management and task scheduling method to realize the spatial data storage and scheduling. As a result of the limitation of performance, relational database already cannot satisfy the user's high concurrency read/write performance and large scale spatial data organization and management. In this article, through debris type partition algorithm for the image division, put forward the non-relational database of remote sensing image data storage strategy based on the Mongo DB, on the premise of high concurrent access, without significant decline of read/write speed. Through experiment contrast the storage efficiency and multiple concurrent access efficiency with the remote sensing image storage method based on SQL Server database, the results show that the multiple concurrent access efficiency of the remote sensing image data storage based on Mongo DB is much better than the latter.