In ideal scenarios, Qlikview follows 3 tier architecture
wherein multiple steps are performed in each tier. Let’s see what this 3 tier
architecture actually depicts.
As shown in the figure above, the first layer i.e. Extract
Layer is used to connect to the data source and fetch the tables. This data
source could be anyone amongst Oracle, DB2, SQL etc. This layer is responsible
for converting the raw tables into the QVDs.
The second layer i.e. QVD Layer aka Transform Layer takes a
single or multiple QVDs generated from the extract layer and apply business
logic to them to create applications specific data. These transformed data is
also stored in QVDs.
The third layer i.e. Data model layer creates perfect data
model by using the QVDs from the transform and the Extract layers. This data
model can be either snowflake or star schema depending upon the complexity of
the data and the tables. This layer can also use inline load, resident loads or
even the excel files.
Presentation Layer takes the binary of the Data Model layer
and displays the UI based upon the user’s requirements. Application or data
security if any is also applied in this layer. This layer can also be used in
Document chaining. Document chaining is the process by which we chain multiple
QVWs with a single QVW. Binary load only takes the data model, it does not take
any layout information nor any variable created in the frontend. We can have
only one binary load in one QVW. Presentation layer just presents the dashboard
for the visualization and the current status of the metrics and the Key
Performance Indicators (KPI).
0 comments:
Post a Comment