Traditional databases provided storage capabilities and offered limited (if any) data processing capabilities. This means that applications would read raw data from the database and, using application code such as ABAP or JAVA, develop complex data processing code to run on the application server.
SAP HANA Cloud database provides sophisticated, multi-tier data storage and also, advanced data processing capabilities so that instead of an application requesting raw data from the database, it can request information. Data processing is now pushed down from the application layer to the in-memory database of SAP HANA Cloud. This means not only is data processing faster, applications are leaner as they do not have to handle data processing logic. Leaner applications means more agility.
Calculation views are defined on top of tables to provide the data processing layer.
Calculation views are usually stacked so that the lower calculation views provide opportunities for re-use by defining only the basic data layer. On top of those are more calculation views that add further calculations and semantics until they are ready for consumption by clients.
Calculation views do not persists data but calculate results on-the-fly based on live data in source tables. Source tables can be local tables in the SAP HANA Cloud database, or remote tables in any database.
A graphical editor is used to define calculation views.
Graphical calculation views are converted into optimized SQL code at run time. The exact generated SQL code depends on the requirements of the query that is calling the calculation view but advanced pruning is applied to create the most efficient and high performance query.
Calculation views allow the modeler to turn data into information by applying data processing function such as aggregation, filters, ranking.
Data from multiple sources can be combined using joins, unions and intersections. The modeler can generate new columns based on any type of calculation.
Data can be organized as a hierarchy to provide drill-down navigation possibilities.
When the standard functionality of calculation views is insufficient, SQL code can be added to provide custom data processing logic.
Input parameters can be defined to prompt the user to provide missing values during run-time, such as filters or calculation values.
SAP HANA Cloud includes advanced data processing engines that combine with calculation views to natively store and process spatial, graph, hierarchy, and predictive data:
- Develop predictive models using in-built algorithms including machine learning.
- Store and query highly networked data models, such as supply-chains or on-line communities (for example, LinkedIn contacts).
- Query hierarchies to extract valuable semantics such as ancestors, dependents, distance between nodes, siblings.
- Store and query geometric data to add spatial information to analytics such as mapping.
With SAP HANA Cloud you can combine any type of data : text, geometries, IoT data, etc to create multi-models that power high-performance, advanced analytics.