Last month, we were honored to receive Oracle Spatial’s Partnership Excellence Award. Oracle supports a wide array of spatial data types, and our goal is to make it easy to extract, transform, and load this data to or from the formats, systems, and data models important to you.
Let’s take a look at a few of the spatial data types available in Oracle:
- Oracle’s SDO_GEOMETRY type is the most flexible and stores both 2D and true 3D (solids and surfaces) vector data. It is also extensible with custom types, e.g. as used by GeoMedia’s oriented points and text.
- Prior to SDO_GEOMETRY, Oracle supported storing geometry in a relational model (i.e., using multiple tables and no special geometry type). This approach was supported up to Oracle 9i R1 inclusive.
- The SDO_TOPO_GEOMETRY type, along with its companion tables, store geometry topologically (i.e., as a set of nodes, edges, faces, and the relationships between them). This allows geometry to be created, edited, and queried while maintaining consistent topology.
- The GeoRaster type and associated types and tables store raster data, e.g. georeferenced images of the earth.
- Point clouds store large aggregates of points, e.g. as produced by LIDAR.
- Triangulated irregular networks (TINs) store large, triangulated surfaces, e.g. the elevation of part of the earth’s surface.
Orthogonal to the data types, but worth mentioning, is Oracle’s Workspace Manager. This allows for concurrent access to different versions of the data, e.g. to support “what if” scenarios or longer duration edits.
Today, FME reads and writes geometry (as either SDO_GEOMETRY or with the relational model), reads topology, reads and writes raster data, and can access versioned tables. You can use FME to indirectly produce Point Clouds and TINs by writing data tables and then using SQL to produce the final objects; for an example see here. Looking to the future, we’re excited about supporting textures on Oracle’s 3D geometries, as well as adding support for other spatial types such as annotations.