“EMCworld is all about transformation – of our customers, the technologies, and ourselves,” says Brian Gallagher, president of EMC’s recently re-named Unified Enterprise Storage Division. One of the major transformations going on inside EMC is that the walls between divisions are being broken down and technologies are being shared.

“We derive the greatest value from complete solutions for individual customers,” he said in an interview on the SiliconAngle Cube from EMC World 2012 (see full video below). “So where we’re going is the integration of our technologies into complete, comprehensive solutions to take a lot of the challenge out of IT.

RecoverPoint, one of the products in his division, is a good example. It came into EMC as an acquisition and was initially added to Invista to provide replication services. Then the RecoverPoint splitter technology was integrated into VMX. Then it was integrated into VMAX 10K. “With 5876 code we now can do that to the 20K and 40Km so we can replicate across the family. And with VPLEX integration we can replicate between EMC and non-EMC storage.

“That is much more beneficial to our customers. Some follow multivendor strategies, others do not, but it gives them the benefit of choice, which is what EMC is all about.”

This supports how customer data centers actually work, he says. Over several years, hundreds of customers around the globe have allowed EMC to analyze how their applications access data. “We collected a ton of metadata about this, over 2 petabytes of trace data, and analyzed over 24 billion IO transactions into storage.”

What EMC learned was that the commonly held belief that tier 1 applications use only tier 1 infrastructure is not true and actually reverses the reality. “The center of gravity in the data center is data,” he said. “And the data is not necessarily mapped to any specific application.” As a result, tier 1 applications often access data sets on tier 1, tier 2, and tier 3 storage.

Normally the value of information declines over time, which is the insight behind fully automated data tiering. But with big data that is not always true, he says. “Depending on the correlations that are made, we see that value of information start to increase as more and more data can be aggregated and correlated. So you can have data dynamically change through the infrastructure, and as you look at those infrastructures it’s less about tier 1, tier 2, tier 3 applications; its more about the data. So we’re driving technology to make it easier to move between our architectures.’

The next level of internal cooperation will be to create internal sources of technology across internal organizations, “our own internal open source opportunities…. We’ve broken down the walls across the organization and have very good collaboration between Isolon, the Unified Enterprise Storage Division, and our backup and recovery division.”