|Oracle9i OLAP Developer's Guide to the OLAP DML
Release 2 (9.2)
Part Number A95298-01
Aggregating Data, 7 of 12
Typically, you will general precalculated aggregates in a batch window as part of maintaining the data in your database. If you wish, you can use Job Manager to schedule batch jobs in Oracle Enterprise Manager, as described in the Oracle9i OLAP User's Guide.
AGGREGATE command aggregates the data for one or more variables according to the specifications provided in the aggregation map. The basic syntax of the
AGGREGATE command is:
variables is the name of one or more variables.
aggmap is the name of the aggregation map.
Your batch job should include commands like the following:
ALLSTAT POUTFILEUNIT=FILEOPEN('userfiles/progress.txt' WRITE) AGGREGATE sales units USING gpct.aggmap UPDATE COMMIT FILECLOSE POUTFILEUNIT
RELATION command only aggregates those source data values (that is, those values that are loaded into the analytic workspace and used as the basis of aggregation) that are in status. The parent values are calculated regardless of whether they are in status or not. For example, if only
MAR01 are in status for the
time dimension, then
Q1.01 will be calculated (but no other quarters), and
2001 will be calculated (but no other years) using only
Q1.01 as input since the other quarters are NA.
This can be useful when you want to aggregate just the new data in your analytic workspace. However, you must exercise some care, as described in "Performing Partial Aggregations".
You can monitor the progress of an aggregation by setting the
POUTFILEUNIT option. You can use the
OUTFILEUNIT option or the
OUTFILE function to set the value of
This command sets
POUTFILEUNIT to the file unit number of the current outfile, which is usually the screen:
This command opens a file named
progress.txt in the
userfiles directory alias, and sets
POUTFILEUNIT to the file unit number of
When the aggregation is complete, you must close the file with a