Databases Reference
In-Depth Information
data layers. The data virtualization process completes the resolution of the contexts and the
relationships, and this can be an automated task reducing room for errors.
●
Data governance
. The data virtualization exercise needs a strong data governance process. By
providing the business users an accelerated solution, the governance process provides a strong
data life-cycle management process.
●
Taxonomy/ontology integration
. Data virtualization can create intelligent links with its metadata
creation and integrate a wide range of discovery and navigation paths by incorporating taxonomies
and ontologies into the solution. This integration also creates a powerful semantic layer to mine
data outside the corporation.
These features of the data virtualization platform will complement ETL and MDM processes and
provide a scalable data integration architecture delivering data to the users in an agile and sustained
process.
Increasing business intelligence performance
Another feature of the data virtualization platform is an increase in business intelligence perfor-
mance. The common processes for BI across any organization include:
●
Data acquisition
●
Data discovery and contextualization
●
Data movement to the EDW
●
Data customization in semantic layers
●
Data presentation to consumers
●
Data personalization for consumers
The biggest issue that remains unsolved is the integration of the data in other business systems,
such as financial, ERP, SCM, and CRM, along with the BI data. This is the biggest weakness that
every user faces in a decision-making situation, and this is the real value of a decision support plat-
form. Data virtualization provides an integrated ecosystem for data integration, and this platform out-
put can increase the performance of the BI solutions out of the box.
Workload distribution
The last architecture advantage of the data virtualization platform that we will discuss is the optimal
utilization of the infrastructure. Data is not physically moved around in the data virtualization architec-
ture; rather, it is stored in the native physical layer. The advantage of this integration is we can tune each
database or data store to serve the native requirements of that platform, and those optimization rules can
benefit the data when accessed from the data virtualization repository. This will enable optimal usage of
the underlying infrastructure and leverage all the resources available within the infrastructure.
Implementing a data virtualization program
Implementing a data virtualization platform needs planning of the data architecture and an under-
standing of data relationships across systems within the organization. Once the data mapping and
