Big data processing platforms such as Hadoop and Spark enable many of today’s consumer and business applications, from Amazon and Netflix to ad targeting systems. But the proliferation of data from cloud applications, the Internet of Things and other sources has also created a monster.

“The Hadoop ecosystem is a zoo of different projects that makes it horrendously complex to administer and build applications on,” said George Gilbert, big data and analytics analyst at Wikibon, the market research firm owned by the same company as SiliconANGLE.

It’s particularly complex to manage data, because the same data needs to be moved and copied to potentially dozens of analytics tools and applications. The process can take up to two years to set up properly and increases the cost to maintain data centers. Public cloud services such as Amazon Web Services and Microsoft’s Azure make data analytics simpler for companies to use, but the same setups at the cloud companies mean more data storage expense and sometimes slower access to the data.

One two-year-old startup is attempting to untangle the current snarl of methods to store and move all that data. Named after the massive Iguazu Falls on the border of Brazil and Argentina (pictured above), connoting the volume, velocity and variety of big data, Iguaz.io today is introducing a conceptual framework it calls the world’s first virtualized data services architecture.

What does that mean? Essentially, Iguaz.io proposes a sort of software orchestra conductor that consolidates data into a repository and makes it available to many applications as files, messages, streams and other forms. Using what it calls data containers, Iguaz.io’s system allows the data to be stored only once and then accessed through application programming interfaces in the form each application needs. “We do transformation of the data on the fly for the various services,” Yaron Haviv, Iguaz.io’s cofounder and chief technology officer, said in an interview.

Haviv contends that the company’s architecture will be 10 to 100 times faster than alternatives, because it bypasses bottlenecks in current operating systems, networks and storage systems. He claims it also provides better data security, a key factor for data that’s shared among many applications and users.

Image: Iguaz.io

“There’s a gap for the new age of applications,” said Haviv, whose Israel-based company is introducing the architecture at the Spark Summit conference in San Francisco. “We’re going to fill this black hole in the enterprise.”

Right now, Iguaz.io is not yet selling a specific product. Rather, it’s pitching an architecture or conceptual structure. But eventually, said Haviv, the company could provide software on a computer appliance for on-premises data centers. It’s also working with unnamed “second-tier” cloud providers to offer its software.

Iguaz.io may compete with, among others, MapR, which has a set of storage engines to simplify data management, a different approach to the problem. But it also will compete with public cloud providers. Either way, it has to persuade large enterprises to take a flyer on a young, unproven company, albeit one headed by experienced technologists.

Iguaz.io, cofounded by Haviv and Chief Executive Officer Asaf Somekh, has raised $15 million in venture funding from Magma Ventures, Jerusalem Venture Partners and some strategic partners. Somekh and Haviv were executives at networking equipment maker Mellanox Technologies, which in 2010 had bought Voltaire, the networking firm where they also worked together.

Photo from Parque Nacional Iguazu