The Logical Data Layer with Zero Replication
Data virtualization is a logical data layer which enables multiple siloed
databases as well as data sources to be accessed and viewed as a single
database. Unlike traditional data integration, where physical ETL is
necessary; data virtualization servers perform data extract, transform and
integrate virtually.
In this process, there is zero replication where there is no need to move or
copy data from the source systems. The modern-day analytics depend
upon distributed data systems and a variety of data to produce insights. In
such a scenario, data virtualization holds the key as it can ensure a unified
‘single source of truth’ without the necessity of the cumbersome process of
data extraction and load.
This eases out the overall turnaround time of any use case delivery. With
the virtual data integration approach, Data virtualization does not require
large infrastructure, thus saving the Capex for such initiatives. Also, with
the change in data at data sources, the same will be reflected in the
analytics use case due to the data virtualization framework.
