Show TOC

Multi-Temperature Categories and BW Architecture (LSA)Locate this document in the navigation structure

The various areas in a Data Warehouse and the various layers of the BW architecture (Layered Scalable Architecture - LSA) can be assigned to the following multi-temperature data categories: hot, warm and cold.

The following illustration shows how the data is distributed between the different areas of the BW architecture after the data is classified as hot, warm and cold:

The following sections provide you with further information about the multi-temperature data classification, how this classification is used in the BW architecture and the solutions recommended by SAP for multi-temperature data management.

Data category: hot

How frequently is the data accessed? This data is accessed very frequently for reporting or by Data Warehouse processes.
How is data used? The EDW and Architected Data Mart Area (Reporting Area) contains data, which is frequently accessed by BW HANA and SAP HANA functions, for example DataStore object activation processes, forecasting functions or queries. In a BW system, this data is usually located in DataStore objects (advanced), Standard DataStore objects, DataStore objects for direct updates or in InfoCubes.
Solution The data is in the main memory of the SAP HANA database.
Features There are no functional restrictions.

Data category: warm

How frequently is the data accessed? Generally this data is only accessed rarely. This data does not have to be permanently stored in the main memory.
How is data used?

The acquisition area or input area receives data from the source system. Here the data is accessed once, in order to prepare and harmonize the data. Then the data "waits" here to be either deleted or archived. Data is only deleted or archived once a specific number of days, years or months has elapsed and once there is confirmation that the data has been stored (harmonized, prepared, transformed) in higher layers (such as the reporting layer). In a BW system, this warm data is usually located in persistent staging area tables (PSA tables) of BW DataSources.

The complete history of the loaded data is stored in the corporate memory area. This area can serve as the source for reconstructions, without the need to access the source of the data again. Some Internet companies, for example, retain their clickstream history in this area for a period of 10 to 15 years. In a BW system, this warm data of the corporate memory is usually located in DataStore objects (advanced) or write-optimized DataStore objects.

Recommended solution SAP HANA SAP HANA Dynamic Tiering for Using Extended Tables
Note Note the licensing conditions when using SAP HANA dynamic tiering.
Features There are no functional restrictions.

Data category: cold

How frequently is the data accessed? The data is no longer needed in the BW system. It is only accessed sporadically. This data does not need to be saved in the SAP HANA database.
How is data used? After a specific time, the data retention period expires for data stored in the hot data area. However, this data may still need to be accessed sporadically. For example, data from the last two years must be stored in the hot data area, whereas older data must be stored in a cold data area outside of SAP HANA. BW data archiving processes write this cold data to a near-line storage solution (such as SAP IQ).
Recommended solution SAP IQ as a Near-Line Storage Solution
Features Restricted to (BW) NLS. The data can only be accessed with read access, for example.