Design Question

[mce]

Hi,
I am new in TM1 and have the following design question for the experts:

Suppose I have a Price cube where I have prices by year and by product. and suppose I have a Revenue cube that I calculate Revenue based on quantity (multiplied by price from price cube) by month, by product and by client.

I have 2 options to calculate revenue:
1- without having Price in my Revenue cube, using DB function in the multiplicaton formula.
2- having a Price element in Revenue cube that is pulled from Price cube using a DB function and then calculating revenue based on quantity and price in the same cube without any DB function in the multiplication formula.

Basically in the second option we do populate price for each and every client and month before using it in a revenue calculation, while in the first option we do not populate price in Revenue cube but pulling it from price cube for each and every calculation.

The question is which option performs better in TM1 in terms of speed of calculations and in terms of memmory usage?

Thanks in advance for all replies.

Regards,

[Jeroen Eynikel]

Using a DB formula rather than pulling it in is obviously the better solution.

It is a step less in the calculation and it will require less memory as if you also pulled them in first you would require additional feeders.

[jim wood]

I agree. Having the price field in the revenue table will also lead to repition fo data. (Not just between the 2 cubes)