hello gurus,
I am facing a lot of performance related problems. hence decided to optimize the system by cleaning the unneccessary objects.
my opinion is that removing the unwanted characteristics relationship will certainly improve the performances.
my current planning cube has the foll. characteristics relationships defined on them
No Source Char -
Target Char -
Usage
Type
1 Company code, Controlling area,Currency key----
na -
without derivation -
Attribute
2 Version, Value Type for Reporting -
na -
without derivation -
Attribute
3 Company code -
Controlling area----
with derivation Attribute
though i have many other char relationships also which are essential , hence i havenot touch them
however, for the above ones , i feel i can remove them .
eg. co area is defined as a attribute of the Company code , hence i feel the query would anyways pick up the corr right value for Co area .. the relationship between COA : CC = 1:1.
or even company code to currency can be a better eg.
so the question is
if a characteristic is already defined as a attribute in RSA1& they have , do we still need a characteristics relationship between them ?
i feel increasing the no of such relationship will only worsen the performance
give your suggestions .
thanks
ramesh