Spatial partitioning of tables

Tables with geometry can also be partitioned with a spatial key that is, the partitioning is done such that the rows that are close to each other in space will likely be placed in the same partition. There are several methods for using the spatial key to partition the data, and these methods vary depending on the type (point, line, or polygon) of spatial data. One of the main advantages of using a spatial partitioning key is the spatial pruning that comes into play during the query execution. In this section, we explain this concept and show how spatial pruning can help reduce query runtimes. First we look at tables with point data and explain the spatial partitioning of tables.

Single column key

Range-based partitioning requires a scalar (number, character, date, and so on) value as the partitioning key. From the geometry value, we can use either X or Y as this scalar value as the partitioning key. If the table has non-point geometry data, we need to map the...