Managing massive remote sensing data in a database: Using a pure object database as a solution

摘要

The Data Rod model is an object-oriented approach to remote sensing data management that provides the ability to query grid cell changes through time. Grid cells are the atomic data structures within our pure-object databases. A data rod is the time-series collection of data at a grid cell location, and can be visualized as a record of infinite length that contains all the information known at each grid cell through time. This resolves the long-standing problem of managing time-series data and opens new possibilities for temporal analysis. Our initial databases are extensible to any field of study with spatially referenced data. Several prototype databases have been created that contain greater than 2 billion grid cell objects. Current development efforts include the creation of a database containing the complete 5km AVHRR satellite data set (28+ years) covering all of Greenland.