3 Aggregation

In EQL, aggregation operations bucket a set of records into a resulting set of aggregated records.

• GROUP/GROUP BY clauses
  The GROUP and GROUP BY clauses specify how to map source records to result records in order to group statement output.
• MEMBERS extension
  MEMBERS is an extension to GROUP BY that allows grouping by the members of a set.
• GROUPING SETS expression
  A GROUPING SETS expression allows you to selectively specify the set of groups that you want to create within a GROUP BY clause.
• ROLLUP extension
  ROLLUP is an extension to GROUP BY that enables calculation of multiple levels of subtotals across a specified group of attributes. It also calculates a grand total.
• CUBE extension
  CUBE takes a specified set of attributes and creates subtotals for all of their possible combinations.
• GROUPING function
  The GROUPING helper function indicates whether a specified attribute expression in a GROUP BY list is aggregated.
• COUNT function
  The COUNT function returns the number of records that have a value for an attribute.
• COUNT_APPROX
  COUNT_APPROX returns the most frequent refinements.
• COUNTDISTINCT function
  The COUNTDISTINCT function counts the number of distinct values for an attribute.
• APPROXCOUNTDISTINCT function
  The APPROXCOUNTDISTINCT function counts the number of distinct values for an attribute.
• Multi-level aggregation
  You can perform multi-level aggregation in EQL.
• Per-aggregation filters
  Each aggregation can have its own filtering WHERE clause. Aggregation function filters filter the inputs to an aggregation expression. They are useful for working with sparse or heterogeneous data. Only records that satisfy the filter contribute to the calculation of the aggregation function.