Hello All,
I am reading to use Union with constant values when dealing with multiple analytic views (so each analytic view has it's own unique measures). In our example though a union alone wont' work because dimensions (charactersitics) are also different in each analytic view. So really a join is necessary. Does a union with constant values solve this at the calculation view? Meaning it will combine into 1 record like a join? We don't think so, but want to make sure. Documentation states that joins will have severe performance (slow). But unions won't work. The final output are characters and measures from multiple tables. So:
1. Does Union with constant value work here (we think this only means to set the measure to constant 0, but each analytic view must have same identical characteristics in output. If so, this doesn't work.
2. If Union with constant value doesn't work then should we do all these tables joins at the analytic level (1 analytic view with many tables joined there in the foundation) or still have multiple analytic views and join at the calculation view level?
The problem at calculation view level (graphical) is that a join can only have 2 inputs. So if we have 4 indiviual analytic views these "joins" turn into a "pyrmaid". So we are not sure which is better for performance.
Thanks!