Group operations aggregate data over multiple rows. We discussed the GROUP BY clause and basic group operations in Chapter 4. Decision-support systems require more complex group operations. Data warehousing applications involve aggregation over multiple dimensions of data. To enable effective decision support, you need to summarize transaction data at various levels. We discuss advanced group operations used by decision-support systems in this chapter.
Oracle provides several handy SQL features to summarize data. These include the following:
A ROLLUP function to generate totals and subtotals in the summarized results.
A CUBE function to generate subtotals for all possible combinations of grouped columns.
A GROUPING SETS function to generate summary information at the level you choose without including all the rows produced by the regular GROUP BY operation.
The GROUPING, GROUPING_ID and GROUP_ID functions to help you correctly interpret results generated using ROLLUP, CUBE, and GROUPING SETS.