Cerebiz
HomeContact
 Product » Architecture
 
Most Organizations have disparate data sources such as client/server databases, legacy mainframe systems, and stand-alone file-based systems. We transform and transfer all your business information into a Data Warehouse that is unified, subject oriented, and granular.

The process of Extraction, Cleansing and Loading data into a Data Warehouse is one of the major activities in the Data Warehouse life cycle. This process involves Data Validation, Migration, Scrubbing and Transformation.

Data Validation is a process that ensures the validity of your business information. Your business information should be validated before transferring to Data Warehouse because it is considered as a Read-Only data source and it is optimized for data. Data Migration involves moving data from the source, either directly to the target database or into an intermediate database. Should the data be migrated to an intermediate database, Scrubbing is necessary to ensure that the data is consistent. Transformation is a process which changes data prior to transfer of data to the data warehouse. Combining data fields into one and/or breaking one into many are some of the changes required.

Building cubes for a Data Warehouse and Data Marts is the next step of the process. Prior this step, identification of dimensions and Key Performance Indicators (KPIs) is necessary. The process of identification is based on business drivers and business objectives of an organization.

Once the above tasks are completed, it is time to analyze the business information that is centralized. The basic requirement of the data warehousing concept is the “viewing of data at different levels of granularity – also called drilling-into”. Most common drilling methods are drill- down, roll-up, slice, and dice. Another requirement of data warehousing is Data Mining. Discovering hidden trends / patterns within the business is the primary objective of Data Mining.