Data Virtualization involves separating the physical source of data from its potential applications, with the goal of increasing business agility. It's a twist on the role of master data management (MDM), which is a must whenever one works with datasets of a certain size or diversity.Instead of physically grouping and storing datasets together -- as in one big giant database or data warehouse stored on a big wonking server or cluster -- virtualized data is stored in a distributed manner, across a series of disparate servers.
