- The data is handled through in-memory technology in the working space
- The Tables are stored in a column-based manner
- The secondary indexes are not needed anymore
- The new database is based on code-to-data paradigm
- Value compression
- In-memory technology
Neue ABAP-Function for Code-to-data paradigm.
- CASE Statement
- UNION Funktion
- CDS Views on the application server
- SAP List viewer with integrated data access (showing the data directly from the database)
The Server today is costly due to the performance issues and large memory space is needed for the RAM.
SAP HANA is used to save the memory space by bringing in-memory technology.
Persistence view is used to hold the data in following cases.
- When the data occupies more space in the hard disk.
- When the server is down
- The changes in the data records will be saved in the logs. Those changes will be updated on the HANA database and the buffer will be deleted.
- The changes in the column-based data tables cannot be saved directly. The data has to be organised with the help of temporary tables, before it is saved to the column based tables.
2. SAP HANA Column based approach:
Example Data table:
| Food item | Food Id | Food cost | Currency |
| Cool drink | 100 | 20 | INR |
| Cheese burger | 101 | 30 | INR |
| Ice cream | 102 | 15 | INR |
- Column based approach:
Both row based and column based approaches are possible in SAP HANA .
Column based approach: This approach is useful for aggregation purpose. For example in order to find the total cost of the food items, the column based approach is useful as the total can be determined in a single line.
- Row based approach:
This approach is useful for transactional data. For example in order to change the currency of a particular food item.
In the column based approach, when we insert a single line ( One column ), the table needs to be reorganised before it is saved.
The column based and rows based selection options are available from SAP Netweaver ABAP 7.4 version through ABAP Dictionary (SE11 Transaction).
3. Value compression
In the column-based approach, there is only one data type for a series of values in the column. It makes it easy to compress via attribute vector.
A compressed column will be created with every value and a numerical id is assigned to it. An inverted index table is created with the index ids and lines.
| Inverted Table | |
| Index | lines |
Working with numbers is easier than strings. String operations takes a lot of load on the central processing unit.