Better Data Model

[#1]

Hi, i would like to know as many opinions as possible about, what is the best approach to create data model in QlikView. What is better for QlikView performance?

1. Concatenate different fact tables into one big fact table and after that join all dimensions to this table and have a nice star schema. Add source column, to distinguish different data sources. Do set analysis to select from big fact table only those fact that belongs to first resp. second fact table, that are concatenate now.

2. Load each fact table as separate table in data model and connect shared dimensions trough linked table and other specific dimension directly to appropriate fact table.

I am asking this, because i am about going to build QlikView app with various sources, planned is 5 different sources with data about same business, but different parts of it, dfata are barely connectable and there is different granularity. This fact tables have many shared dimensions like country, customers, product and so on, but each one has few specific dimensions. So before i start, i have to find out best approach to have model, that is readable and suitable for to reach best performance.

Thanks

[cheburashka]

Hello Jozef,

I guess you already finished your datamodel. But should you be interested in some tools to help checking your datamodel while building it. I created a "Key check module" (a set of objects and a script subroutine) to check the links between tables. You can read a short post about the "Key Check Module" here.

Hope you find it useful.

Have fun Qlik'n, Koen

QlikViewModules.com