Aggregates

Aggregates:-
-----------
* Smaller Cube built on a Info Cube to Improve the Query pderformance.
* If we dont rollup the request it will not be available for reporting.

* Activate / Deactivate
* Switch on / off
* Delete

Aggregates:it is a subset of an info cube---baby cube---in the aggregates the data is stored in aggergated formate

icube
eid/calmonth/calquarter/sales
100/01/98///1/1000
200/01/98///1//1200
100/02/98///1///3400
200/02/98///1///4000

q1----eid/quarter/sales

db---records---4////olap---4 recordrds----------eid/quarter/sales
    100/1/4400
    200/1/5200



aggregate---calyera qurter/sales
  01/9600


q1---quarters/sales




ic1---aggr1/aggr2/agr3-olap proces----db read time///olap agrgation time


icube---detailed

aggergated---summrized data-----duplication of data-------

rsrt-------query trace


types of aggrgates:

1.aggregate by all char values---aggergate the data wrt all char combinations

calyeraquarters///sales
01/98////1000
01/98///2000/
02/98--300
04/98---5000
03/98---1200
03/98---1800

aggregates all chars

01/98---3000
02/98////300
03/98////3000
04/98///5000


2.aggergate by fixed value--single char value


country/sales
us///3000
us///5000
ca//2000
ge//4000

aggr---us
us/8000


3.aggregate by hier level
 level-1/2/3/4/5---each level we can maintain

1.aggregates----Mantain aggregates
aggr--simial r to info cube str-----fact tab---dim table---3 fixed dim--13 user dim

aggregate---fact table---2 fixed dim(unit/data package)+char(dim)


NOte:while loading data to info cube switch off the aggregate///aftr data loading switch on the aggregate

aggr--us

r10---ss---us/ca/gb-----1000---us---100

loading data to infocube and aggregate parally tampers ur loading preformance--


loading data from icube to aggregate----rollup