Skip to main content
Announcements
Qlik Connect 2024! Seize endless possibilities! LEARN MORE
cancel
Showing results for 
Search instead for 
Did you mean: 
Not applicable

Multi-tier Architecture

Hi, Trying to build a small app. Not sure if it works, but can I have a data layer (in Excel), QVD extraction layer (Excel to QVD file), QVD config layer (Excel with formula or chart specification), and finally Application layer (presentation)? Would help if someone could clarify. Thanks.

1 Solution

Accepted Solutions
MK_QSL
MVP
MVP

Source : QlikView 11 For Developers Book

Two-stage architecture

The two-stage architecture is composed of the following layers:

Source Layer: composed of the source databases and original tables.

Extract Layer: composed of QlikView documents, containing mainly script. These

are used to pull the data from the source layer and store it into QVD files. The

extraction scripts can either create a straight copy of the source tables to store

them into the corresponding QVD files, or perform certain transformations before

storing the result.

QVD Layer: the set of QVDs resulting from the Extract Layer. These QVDs become

the data sources used by the final QlikView document.

Presentation Layer: the set of QlikView documents used to provide the data to the

end user. These QlikView files will use the QVDs created in the previous layer as

data sources, and sometimes perform additional transformations to create the final

data model. No database calls are performed from the presentation layer.

Three-stage architecture

In this architecture, two additional layers are added: Transformation Layer and

Transformed QVD Layer. The role of these two additional layers is to hold all

transformations that need to be performed upon the source data before it can be integrated

into the target data model.

View solution in original post

4 Replies
Gysbert_Wassenaar

That's certainly possible. If it makes sense for a small app is something you have to decide for yourself. You might find the QlikView Deployment Framework interesting.


talk is cheap, supply exceeds demand
MK_QSL
MVP
MVP

Source : QlikView 11 For Developers Book

Two-stage architecture

The two-stage architecture is composed of the following layers:

Source Layer: composed of the source databases and original tables.

Extract Layer: composed of QlikView documents, containing mainly script. These

are used to pull the data from the source layer and store it into QVD files. The

extraction scripts can either create a straight copy of the source tables to store

them into the corresponding QVD files, or perform certain transformations before

storing the result.

QVD Layer: the set of QVDs resulting from the Extract Layer. These QVDs become

the data sources used by the final QlikView document.

Presentation Layer: the set of QlikView documents used to provide the data to the

end user. These QlikView files will use the QVDs created in the previous layer as

data sources, and sometimes perform additional transformations to create the final

data model. No database calls are performed from the presentation layer.

Three-stage architecture

In this architecture, two additional layers are added: Transformation Layer and

Transformed QVD Layer. The role of these two additional layers is to hold all

transformations that need to be performed upon the source data before it can be integrated

into the target data model.

Not applicable
Author

Thanks. Looks like a restricted group, awaiting acceptance.. Meanwhile, is there any other link you could share? Thanks again.

Michael_Tarallo
Employee
Employee

Hello Rakesh - this greatly depends on how big of an application / deployment you are trying to build and what your current knowledge of QlikView is. Also depends on what components you are licensed for. It is possible that QDF might not be necessary. QDF helps with structure, management and organization of a QlikView deployment and is usually seen in larger deployments. By all means check it out if you wish.

What you described is absolutely possible and configurable amongst different physical and logical  tiers.

  • QlikView Server / Engine - calculation, in-memory engine
  • QlikView Web Server (default) or IIS - entry point for presentation layer (AccessPoint)
  • Distribution Service (Publisher) - scheduled tasks, reloads, data reduction, distribution
  • Directory Service - connects to various security providers / windows authentication by default

Note that QVD files are simply staging areas (created as files) that represent the uncompressed version of the QlikView in-memory data table model. They are used to create a semantic-like layer or data-mart that is shareable and reusable with all data, calculations, expressions, common field names etc stored. They are also used for loading data very quickly into the QlikView application memory vs. using the standard SQL extract and LOAD in a QlikView application. So you can extract from Excel, transform and calculate with QlikView LOAD script and then STORE the table into QVD for later use and reuse. The QVD files can then be used to create and load the QlikView applications.

Your QlikView application (visual and presentation) can be segregated - you can have two separate apps (.QVW files). One that extracts the data (using the LOAD script and functions etc.) from excel, transforms and creates the QVD. And one that simply just reads the QVD, loads the data and has all your visualizations, selection objects, boxes etc for analysis.

You can take the first applications (ie: .qvw 1)  and schedule it using Tasks in the QMC (QlikView Management Console) to be reload with the QlikView Distribution Service (Publisher). This will create, update, overwrite, append (depending on your load ogic that creates the QVD - it varies) - the QVD file so the application with the visuals has fresh / updated data when re-loaded.

Please note that all of this can be done directly from one QlikView application without storing the data into a QVD file. QVD files are recommended for a number of reasons, but without any of that knowledge I can only make some assumptions.

Applications created with QlikView Desktop can be simply copied to the User Documents directory that you have configured in the QMC (requires knowledge of understanding Source Documents and User Documents folders and if you have the Distribution Service (Publisher) installed. Also depends on what security you are using, Windows folder security or QlikView Security. Once all configure properly - your web browser accessing AccessPoint at : http://hostname/index.htm will display the application and can be access over the web in the same manner that our demo apps are:  http://demo.qlik.com

Hope this information is helpful:

Michael Tarallo

Senior Product Marketing Manager

Qlik

@mtarallo - follow me

Regards,
Mike Tarallo
Qlik