A major feature of the ISO/ANSI SQL extensions for OLAP is a construct called a window. This windowing extension let users divide result sets of a query (or a logical partition of a query) into groups of rows called partitions and determine subsets of rows to aggregate with respect to the current row.
You can use three classes of window functions with a window: ranking functions, the row numbering function, and window aggregate functions.
Windowing extensions specify a window function type over a window name or specification and are applied to partitioned result sets within the scope of a single query expression.
Windowing operations let you establish information such as the ranking of each row within its partition, the distribution of values in rows within a partition, and similar operations. Windowing also lets you compute moving averages and sums on your data, enhancing the ability to evaluate your data and its impact on your operations.
A window partition is a subset of rows returned by a query, as defined by one or more columns in a special OVER() clause:
OVER (PARTITION BY col1, col2...)
For information on analytical function support of the LONG BINARY and LONG VARCHAR data types, see Funtion Support in Unstructured Data Analytics.