Skip to main content
Announcements
See what Drew Clarke has to say about the Qlik Talend Cloud launch! READ THE BLOG
cancel
Showing results for 
Search instead for 
Did you mean: 
sarfaraz_sheikh
Contributor III
Contributor III

Which Approach is better

Hello Folks,

I have created the data model in three different ways, Can anyone tell me which one is better approach to follow.

1)  Approach


Approach 1.png

2> Approach :

Approach 2.png

I wanted to know which one is better from all respect like performance, reload and etc.

Guidence would be much appreciated.

Thanks

Sarfaraz

14 Replies
pradosh_thakur
Master II
Master II

Haha. You never  miss anything sunny!!


he was using apply map in both the approach so in edition to your answer i posted those link.

one quick query here. If we have lets say few million of records and we do it according to first approach than we will have one fact table to associate with other tables. will it not take less time when compared to having two tables which are associated on run time with each other as well as other tables.

I read in the community that

1: Qlikview association is comparable to  Outer join but at run time and not during script execution.

2: when we compare two tables joined into one table and those two tables not joined and in form of two tables (like join and keep) than the slicing and filtering of joined  table will be faster.

please correct me sunny if i understood anything wrong here.

regards

Pradosh

Learning never stops.
sunny_talwar

Yes, we have seen advantages (slightly faster running expressions) of combining a table into a single fact table with massive amount of data. But I am not sure how advisable it is to actually do this all the time.... the reason i say this is because why not just have a single table model all the time? Because things are not always simple, and at times it is easier to maintain and understand a multiple table model. So, I think it comes down to preference of what your data model requirement need is.

But coming back to the question, between the first two options, I don't see any difference in reload times. On the front end, we might see some differences with massive amount of data, but would be negligible for few million rows.

sunny_talwar

Let me just clarify that the above is based on what we have seen in the past. I have not done any testing and is my opinion. I suggest performing some sort of tests before making final decisions.

pradosh_thakur
Master II
Master II

thank you sunny.

Learning never stops.
qliksus
Specialist II
Specialist II

Approch 2 doesn't have any subfield so that might reduce the load time a bit and with just 2 fact tables and you letting Qlikview create its own join instead of a explicit one using the join keyword i feel approach 2 would be better