Qlik Data modelling with few feilds similar

[sujit_nath]

 have 2 data sets with millions of rows. Suppose Table 1 is HR data and Table 2 is Admin data. The employees in both the tables are different.

But the scenario is that Table 1 and Table 2 has 10 columns, 5 of which are common. Other 5 are different in HR and Admin tables.

How do I model this data? If I simply concatenate, the complete dataa will have 15 columns. and 5 will be NULL for HR and 5 will be NULL for Admin.

How do I address this?

[tresesco]

If the data is not that huge and performance is not a big issue, I would concatenate the tables using a flag field in each table, so that I can use them in set analysis in UI. Otherwise, linked table would be my approach.

[sujit_nath]

Hi Tresesco,

Since the data is huge, performance is an issue here. That's the reason I can't concatenate. Also Link table won't solve my purpose I believe, as there are no common values in the data set. It's just that few headers are same.

[tresesco · Accepted solution]

---

@sujit_nath wrote:

Hi Tresesco,

Since the data is huge, performance is an issue here. That's the reason I can't concatenate. Also Link table won't solve my purpose I believe, as there are no common values in the data set. It's just that few headers are same.

---

If performance is an issue - you should try to model this concatenating the tables. If you are worried about nulls in the table, that should not be a significant performance defining factor. And I don't know why you said linked table won't solve your problem. Have you tried the solutions? Start with concatenating the tables.

Usually these are the ways to resolve such issues in modelling.