Building data marts through code free ETL Tools
When considering tools to construct datamarts, companies have a broad range of selections. Two classes of tools exist: Extract, Transform, and Load (ETL) tools and Data Cleansing tools. When we talk about ETL tools we mention to the old-fashioned technology terms input, process, and output. Irrespective of the marketing point of view, these tools all move data from source to target. While license cost for these ETL tools is high, the benefit/cost ratio is more convincing.
There are some features which are necessary for building datamarts in all the ETL tools :
Database Platform Support: Most ETL tools offer native platform support for the major database platforms. The majors: IBM, Oracle, Sybase, and Microsoft are approximately always supported. Support choices vary for Informix, Ingres, and other tier two players.
ODBC Connectivity: Most ETL tools have directed support for bulk copy programs (BCP) and various ODBC data formats.
Data Integration: ETL tools also offer to support for varied application integration. ETL tools normally support various data formats, including ASCII, EBCDIC, and XML. There is also rising drift to integrate with numerous third-party applications and ERP back office systems.
Data Extraction: Moving data from various source environments into designated targets is a key feature. Business rules are added as data is extracted.
Data Transformation: An important part of the datamart building process is attained during data transformation. At this step conditional and mathematical transformations are achieved. It is quite common to incorporate business related data that is not included in the operational system during transformation. Business rules are added here as well. In this phase logical rules, transformations, substitutions, duplicate handling, and aggregations are performed.
Metadata Management: Metadata management is a key problem that the better tools address to varying degrees. Metadata sharing with third party applications and bridges to business intelligent and OLAP environments are key features to watch for.
Scalability: Because the volumes of data being moved range from small databases to VLDB systemsETL tools typically support load balancing, failover/fault tolerance, and allow parallel processing. These features are also supported to varying degrees.
Development and Administration: ETL tools all have graphical, centralized managerial environments. To various degrees, they have load monitoring, auditing, and job scheduling features.