Skip to main content
cancel
Showing results for 
Search instead for 
Did you mean: 
Not applicable

Re: Qlikview Sense x Tableau

Michael Tarallo

can you help me????

14 Replies
Michael_Tarallo
Employee
Employee

Hi Edson,

Another powerful capability of Qlik products is that they have a easy to learn and use scripting syntax that allows you to perform "Transformations" on data - it is ETL but not in a graphical sense. (it is not always necessary to use or write as the Wizards can also input the syntax for you) The syntax provides a robust library of methods and functions to merge, augment, define and manipulate data, including the use of variables.

A QVD file - is a file based uncompressed in-memory representation of the Qlik in-memory index. It is not always necessary to create a QVD file - but is recommended when implementations grow and centralized storage and re-usability of common expressions, fields and other assets becomes necessary. A QVD can be used to store the data model and the data - and the used to quickly load the in-memory index of a QlikView application or Qlik Sense app. QVDs offer efficient centralized reuse and storage and loading for applications.

Some customers - use the script to combine multiple data sources, normalize and derive the expressions and data columns and then store the data model as a single table into the QVD file. Which is then later loaded on a scheduled basis into a Qlik Sense app or QlikView application.

ETL - whether it is from script an out-of-the-box graphical tool or a partner's solution - will always be part of BI solutions. New data is always being added, data needs to be conformed and governed - and ETL tools assist in this process.

HTH

Mike

Regards,
Mike Tarallo
Qlik
jaimeaguilar
Partner - Specialist II
Partner - Specialist II

Hi,

I've worked heavily with QlikView and just some personal documents with Tableau. According with what I've seen and read in other discussions I can tell you this:

  • Tableau connects directly to data sources and executes queries on the fly, besides I think Tableau has some native connectors which in theory could make connections faster.
  • On the other side, QlikView/QlikSense as an in-memory tool uses their own data files (QVDs) in order to have a replica of you databases and then loading into memory. It is not strictly necessary to store data into qvds before loading it in memory, however it is really useful. It seems that Tableau also has a somewhat equivalent to qvds in TDE files.
  • QlikView has a relatively new feature called "direct connect" which lets you connect to databases live and combine it with in-memory technology. It is still in development but is a step forward for real time analysis.
  • Tableau doesn't have a script engine or an ETL structure, however you can make some processing on the fly, like joining table or make some calculations similar to what Set Analysis does.
  • I think in some way Qlik Sense is a response to the success that Tableau has had in order to have a tool focused on end users and in the idea of self-service BI / Data discovery,

regards

Not applicable
Author

Wonderful James,

you are really master user of both tools.  great points you made on Tableau & Qlik Sense.

Wonderful

Thanks

Michael_Tarallo
Employee
Employee

Hi Eduardo,


You made a good point. Tableau and Qlik are fundamentally different when it comes to working with multiple heterogeneous data sources and blend the data in a visualization. (the blend part is very important)


To read more about how Tableau works, please read this recent blog post on Tableau website: http://www.tableausoftware.com/about/blog/2014/5/9-data-blending-tips-data14de-30845

The way Qlik works with multiple data sources (whether there are 3, 5, 10….30) is no different than when with one single data source. All data is equal in Qlik, there is no primary, no secondary data sets, all data can be analyzed the way it makes sense for the users to visualize and explore. This architecture allows us to work with a greater number of data sources, with large volume, without data loss in a “join” (see point #5, Tableau data blending cannot emulate a full outer join)  - while providing users with a consistent user experience and performance.


Hope this helps.


Regards,


Mike Tarallo

Qlik

Regards,
Mike Tarallo
Qlik
Not applicable
Author

One more difference I observed is,

Development Sequence in Tableau 8.2 is :

Create sheet -> Assemble sheets in Dash Board -> Use sheets/Dashboards in Story slide


Development Sequence in Qlik Sense 1.0.2 is :

Create all visualizations on one sheet (In a way, this is what you do in final step of tableau visualization )