Distributed File System Management Technology Research Based On The Massive Remote Sensing Image Data

Due to the high speed development of aerospace technology and internet technology, the data volume of Remote Sensing Image increases rapidly in corresponding, therefore it raise the difficulty of storage and management for the massive Remote Sensing Image data. The traditional way to manage the massive Remote Sensing Image data is the centralized model, the deficiency is that it is easily to reach a system bottlenecks when the image data volume gets huge, thus it takes up most of the storage space, waste a lot of time and resources as well. Therefore, the researth of manage the massive Remote Sensing Image data in a more effectively way become a valuable and urgency issue. While the appearance of Distributed File System just able to solve this problem, it uses distributed management technology, so brings new techniques and methods for the massive Remote Sensing Image data management.Through drawing on the existing distributed file system to the massive Remote Sensing Image data management techniques and methods, in this article we researth and design a Hadoop based massive Remote Sensing Image data management framework, and describe in detail of image data storage model and the massive Remote Sensing Image data storage technology.The specific contents include the following:In this paper, we analyzed and summarized both domestic and foreign situation for the massive Remote Sensing Image data storage management and the distributed file system Several domestic and foreign mainly distributed file system are introduced, including NFS network distributed file system, GPFS distributed file system, AFS distributed file system, Google distributed file system and Coda distributed file system, and points out their advantages and disadvantages.This papers research the relevant technology of the distributed and Remote Sensing Image data structure characteristics, and the massive Remote Sensing Image data logical structure of organization and block technology, then makes a deep analysis and discussion, and researth on the main technical route of the massive Remote Sensing Image data storage and management way, finally points out the deficiency of the existing technology and the characteristics.This paper introduces the HBase database which based on NoSQL, researth on the massive Remote Sensing Image data pyramid technology and image pyramid build technology,and put forward a index method based on Hilbert curve,and a new kind of image data storage model, it can be applied in the situation of multi-temporal multi-resolution image data, also can store and inquires the massive Remote Sensing Image data information of different period in the same area.In the paper, the method proposed by experiment technology is verified.