3

I want to process my data into Qlikview but i am confused about to process the data through Cube or directly from SQL. Can anyone tell me which gives better performance from cube and SQL? Note: I have millions of data into the database.

Crystal
  • 45
  • 8
  • 4
    *You* are in a prime position to tell which of these options will give better performance for your setup -- why don't you try? If the objection is that you can't wait for millions of rows to process, have it read from a view/views that limit the data so you can get a ballpark figure. Reading pre-aggregated data from a cube is almost always faster, but comes with restrictions, precisely because you're no longer free to aggregate or filter data as you please (as opposed to what the cube offers). – Jeroen Mostert Jan 24 '19 at 11:27

2 Answers2

2

Generally as the volume of data grows, the advantages of SSAS tend to become more apparent than those from using SQL Server as the source. How will the data be used? When it comes to large scale aggregations SSAS becomes very beneficial. SSAS will also force a structured layout, as the relationships are predefined in the cube as opposed to joins. Some additional features that SSAS brings are hierarchical analysis (hierarchies) as well as ease of use with tools such as Excel and SSRS, although it sounds like you're only looking to use Qlikview for this. However, your best option would be to do a baseline for both SSAS and SQL Server in your environment with queries that best represent what would be run when this is implemented, and assess the results from there.

userfl89
  • 4,610
  • 1
  • 9
  • 17
1

From BI tool perspective it doesn't matter as you can connect to both source (SQL is more common but it depends on your expertise). Regarding performance the best strategy is to have separate extract layer and store data incrementally as qvd (for example every night previous day) so performance is not as important with incremental reload as even for big data sets it should be quick.

If your original source of data is SQL in my opinion it doesn't make sense to replicate data in 3 places (SQL, cube and QlikView) better connect directly to source save it incrementally raw data as qvd and then have transformer which will model that data.

Hubert Dudek
  • 1,666
  • 1
  • 13
  • 21