Data Warehouse Service (DWS)

Data warehouses are consolidated databases fed from different sources. These sources usually include databases run by various departments of a company. In the data warehouse, the data from the various source databases is consolidated in a standard format and data model, making it possible to analyze data from different sources.

A typical example of such an approach can be found in retail, where various business units maintain different data sources. In order to gain an insight into sales figures of specific products or in specific regions or subsidiaries, data from various sources and maintained using different technologies needs to be linked. This is where the data warehouse comes in. The Data Warehouse Service enables data mining, i.e. the evaluation of large data sets. In turn, these analyses give users new insights and enable them to make better decisions.

The data sources can be located anywhere, but for the evaluation they are copied to the Open Telekom Cloud (data center in Germany). The Data Warehouse Service supports MapReduce clusters and can use HDFS data. Data is imported from object storage at high speed, and encrypted data transfer is also possible. Furthermore, the cluster supports internal DNS addresses, Java UDF and roles issued in IAM. Time zones can also be configured.
The Data Warehouse Service is billed using the pay-per-use model. To this end, virtual machines and storage capacity are made available which – similarly to the Relational Database Service – already include the required software.