Leveraging Virtualization to Streamline Data Management

As increasing data management challenges emerge for enterprises, virtualization offers a solution to reduce data proliferation and ensure that all consumers can leverage accurate information quickly and effectively.

Suresh Chandrasekaran, senior vice president of Denodo;  BJ Fesq, chief data officer at CIT Group; and Kyle Hailey, technical evangelist Delphix,  examined the pros of data virtualization to streamline management processes during Data Summit 2015 in a session titled, “Increasing Agility through Data Virtualization.”

Data Summit 2015 took place May 11-13 at the New York Hilton Midtown.

Data virtualization is like aggregating many data sources and offering a transparent interface of those many data sources, Hailey said.

There are several big strains on IT right now such as how to quickly and efficiently provide environments to developers and QAs, according to Hailey. Virtualization would help ease this pain.

Self-served data is very powerful but it is necessary to make sure that people are speaking from a single source of the truth, Fesq said.

CIT Group offers a solution called the data services layer, which is not synonymous with data virtualization, but helps with getting information to begin the virtualization process, Fesq explained.

“Beyond integration, it provides usage metering, which is key to knowing who is actually consuming the data, how often, and what they are doing with it,” Fesq said. The solution also provides monitoring for in-flight data movement, which becomes important when enterprises start flowing critical data into it.  “You have to be able to support [your data],” Fesq said.

Other solutions include taking file snapshots. However, some bureaucracy in organizations will make it difficult to being the virtualization process. Additionally, technology itself can slow the process down, Hailey said.

While CIT Group offers a data services layer, Delphix offers its own solution called the virtual data appliance. It connects to the source of data and pulls the copy and streams in the data change, keeping track of the time flow the data changes from the source while having an easy to use interface.

Fesq and Hailey outlined several points as to why enterprises should virtualize their data including gaining value from data, quality monitoring, and governance, among others.

A data governance framework is critical to this process to ensure accountability, Fesq explained, so people are empowered to hone the data and clean things up when they need to. “Typically a good data governance framework is going to include some kind of policy which says, ‘here’s how data should be managed,’” Fesq said.

To access Data Summit 2015 presentations, go here.