Data-Mart

A data mart is a copy of the sub- dataset of a data warehouse (DW), (see below ) is created for a specific organization or area of ​​a particular application or analysis. It can also be understood as a partial view of the data warehouse or non-persistent cache. In practice, existing in a data mart database is in some cases maintained in the long term (see also " independent data marts ").

Reasons for working with a data mart ( " copies " of the data warehouse ) instead of direct access to the data in the data warehouse are based on views ( views):

  • The need for special data structures ( which are not present in this form in DW ) for certain analyzes, for example for multi-dimensional analysis, the so-called online analytical processing (OLAP)
  • Better performance (Performance): shift from computing power to another computer and / or relocation of accesses to a different store and / or in the case of local use less network load
  • Autonomy of the user (eg, mobility, independence from other areas of the organization )
  • Access protection: differentiation from other users or opening for other users ( it will not directly access the data in the data warehouse, but only indirectly via the data mart )

Through the use of data marts can on the particular field ( professionally ), the target system (technical) and the users are ( organizational) largely taken into consideration.

Due to the data marts of the database of the data warehouse is distributed. The thus resulting data marts require the data warehouse, so it is called dependent data marts.

In addition to this data mart concept, there is the possibly historically relevant part of the independent data marts, which detached from a data warehouse exists and allows a view of one or more source systems. This architectural approach runs counter to the idea of ​​integration, data warehousing, as it can lead to many island solutions with time. He comes from a time when it is not the central data warehouse, the data warehouse was.

The data model of a data marts often corresponds to a multidimensional or OLAP structure. The technical implementation of this data model can be as multidimensional database as a relational database or, increasingly, in the memory of the computer.

Data-Warehouse
219445
de