Data governance sets the rules for NetApp’s AI storage agents

Artificial intelligence is turning nearly every enterprise workload into a data workload, and data governance is becoming the test of whether companies can trust autonomous systems with the underlying infrastructure. The question is no longer just where data is stored, but also what can be done with it.

Enterprise data now resides in on-premises data centers, public clouds, and neoclouds, and storage providers are scrambling to connect this infrastructure with production AI. According to Sandeep Singh (pictured right), senior vice president and general manager of enterprise storage at NetApp Inc., adding another silo is the wrong answer. He calls for a single data basis that behaves the same everywhere.

“We know customers will use different locations to find the solution that suits them,” said Singh. “What is fundamental for customers is a consistent data infrastructure strategy that underpins everything so they can get a consistent set of data capabilities combined with a consistent operating experience across environments.”

Singh and Helen Yu (left), founder and CEO of Tigon Advisory Corp., spoke with Christophe Bertrand of theCUBE Research and co-host Rebecca Knight at NetApp INSIGHT during an exclusive broadcast on theCUBE, SiliconANGLE Media’s livestreaming studio. They discussed why consistent data, clear responsibilities and policy guidance must be in place before companies hand over infrastructure decisions to AI agents. (*Disclosure below.)

Data management is shifting to the autonomous control level

Consistency is not just a technical problem. Yu, who advises companies on the use of AI, sees the larger gap in the organizational structure. Teams without trusted data bypass official systems, allowing shadow AI and shadow IT to take hold, she noted.

“It’s not a storage issue. It’s a leadership and responsibility issue,” Yu said. “When teams don’t have consistent access to consistent and trusted data, they actually stop making decisions. They start finding workarounds.”

On the infrastructure side, NetApp combines a consistent data plane with a unified control plane shared by agents and people. With its hybrid cloud tools with autonomous control, delivered through the NetApp console, customers can set policies and boundaries for their entire fleet while agents ensure the infrastructure runs within those boundaries. In a keynote demo, agents discovered a nighttime performance anomaly instead of calling a technician, Singh noted.

“Agents can implement the necessary quality of service rules in real time to ensure that the constraints there are met,” said Singh. “And in this way no one is woken up. The correct result is achieved and there is no noise neighbor effect.” And people are provided with an audit trail to go ahead and follow up on.”

Trust in this model depends as much on accountability as it does on automation. Letting an algorithm decide is not a governance model, says Yu, who urges executives to measure automation by business outcomes rather than remote tasks. It extends data management to machines through the RACI framework: responsible, accountable, consulted and informed.

“Now you need to integrate the machine or agents into your RACI diagram. Who owns what data? Who has access to what data?” Yu said. “And then if you make an exception, who has the right to override the exception?”

Here’s the full video interview, part of SiliconANGLE and theCUBE’s coverage of NetApp INSIGHT:

(*Disclosure: TheCUBE is a paid media partner for NetApp INSIGHT. Neither NetApp, the sponsor of theCUBE’s event coverage, nor other sponsors have editorial control over theCUBE or SiliconANGLE content.)

Photo: SiliconANGLE