The database market has been rapidly changing, so much so that it’s hard to make declarative statements about any part of the process. Still, Oracle makes the claim that it takes decades to build a high-performance and stable database and that has been, for the most part, true. So what has changed to support the creation of so many NPP decision-support databases, first on their own and now on Hadoop?

“What’s changed, really, is the interest in doing it in the open-source model,” said George Chow, CTO of Simba Technologies, Inc. Chow was interviewed by George Gilbert (@ggilbert41), host of theCUBE, from the SiliconANGLE Media team, during Hadoop Summit US in San Jose, CA.

It’s hard to say whether Oracle’s statement about taking decades will hold up or not. Chow acknowledged that it’s reasonable to take a decade because you have to build it, test it, and get it out in the marketplace. But with this open-source movement it’s hard to tell. “I don’t recall whether anyone’s done a measurement of innovation in the open or in the proprietary marketplace,” Chow said.

A convergence workload

The failing of the last generation, according to Chow, is that you have to build a secondary system in addition to your existing one if you want to do analytics. While it’s still in the early stages, however, there are some working to do just such a thing.

“People are trying to marry that workload and run one system that will run day-to-day and do analytics,” said Chow. But as such a concept is in its early stages, even the bigwigs such as Spanner and Prestor haven’t quite made it big yet with a convergence workload. Still, it’s a very exciting time that will be interesting to watch develop over the next three years, remarked Chow.

Watch the full interview below, and be sure to check out more of SiliconANGLE and theCUBE’s coverage of the Hadoop Summit US.

Photo by SiliconANGLE