It’s been said that data virtualization is less relevant with the explosion of big data platforms that can accommodate massive quantities and various types of information. Data experts at Cisco patently disagree.
“With enterprise data warehouses, Hadoop clusters, and cloud-based data sources like Salesforce, everything is still very siloed,” says Peter Tran, product marketing manager at Cisco. “Pulling it all together into a single repository is unrealistic, but you still need access to all that data, no matter where it is.”
That’s where virtualization shines. Cisco® Data Virtualization software augments conventional data integration by combining data sources to optimize query, compute, and network infrastructures. It allows companies to access and query all types of data across the network as if it is in a single place.
“It isn’t a data store, but rather a middleware layer that delivers ‘virtual’ data to the right place,” explains Rick Schreiber, director of data virtualization services at Cisco. “It’s the layer between data sources—which can number in the hundreds or even thousands—and the applications that consume the data, like analytics engines.”
Without virtualization, data is continually replicated as it is moved from one place to another, or as new schema are applied for analytical or reporting purposes. This not only results in data sprawl, but also unnecessary complexity and cost. With virtualization, however, data can be accessed and utilized without being physically moved. This means less formatting, less replication, faster response, and better scalability.
“Data virtualization is a perfect fit for big, complex data environments,” says Tran. “It complements enterprise data warehouses and Hadoop clusters. In fact, it can help align those two environments, delivering data between them and pulling data from them for business-critical applications.”