The rollup and cube functions of DB2

The effect of rollup is to count the name of a grouped column after group by.

BankID as the first column

SELECT    Case        -grouping (Branchid) = 0 Then        branchid        ELSE ' subtotal '    END as Branchid,        case When grouping (bankid) = 0 Then        bankid        ELSE ' Total '    END as BankID,    SUM (balance) from    ebanksummwhere< c12/>workdate = ' 20140320 ' GROUP by BankID,        Branchid  with  rollup

Branchid as the first column

SELECT    Case        -grouping (Branchid) = 0 Then        branchid        ELSE ' subtotal '    END as Branchid,    case< C22/>when grouping (BankID) = 0 Then        bankid        ELSE ' Total '    END as BankID,    SUM (balance)    from Ebanksummwhere    workdate = ' 20140321 ' GROUP    by Branchid,    BankID with rollup

The cube function counts all the columns that follow the groupby. The difference between a and rollup is the column to be counted. Rollup counts only one column, and cube counts all columns

SELECT    Case        -grouping (Branchid) = 0 Then        branchid        ELSE ' subtotal '    END as Branchid,        case When grouping (bankid) = 0 Then        bankid        ELSE ' Total '    END as BankID,    SUM (balance) from    ebanksummwhere< c12/>workdate = ' 20140321 ' GROUP by BankID,        Branchid  with  Cube

The rollup and cube functions of DB2