NetApp INSIGHT 2025: The Enterprise Data Platform—Intelligent, Resilient, Disaggregated | Sandeep Singh, NetApp

Clips
More from NetApp INSIGHT 2025: The Enterprise Data Platform—Intelligent, Resilient, Disaggregated

Sandeep Singh

SVP & GM, Enterprise Storage

NetApp

play_circle_outline NetApp's history of innovation and adaptation to new technological waves.

play_circle_outline Unveiling NetApp AFX and AI Data Engine: Transforming Data Services at NetApp INSIGHT 2025 with Keystone Excellence

play_circle_outline Strengthening Cyber Resiliency: Innovative Data Protection and Rapid Ransomware Recovery Solutions for Modern Storage Systems

Info
Transcript

Sandeep Singh, NetApp

Sandeep Singh

SVP & GM, Enterprise Storage NetApp

In this NetApp INSIGHT 2025 conversation, theCUBE Research’s Christophe Bertrand sits down with Sandeep Singh, senior vice president and general manager of enterprise storage at NetApp, to unpack a “blockbuster” slate of announcements: NetApp AFX and the AI Data Engine. Singh explains why AI in the enterprise is fundamentally a data problem and details how NetApp is approaching AI readiness with secure multi-tenancy, data mobility without copies (while preserving permissions) and hybrid cloud workflows that bring on-prem datasets closer to GPUs. The discussio... Read more

explore Keep Exploring

What has been the evolution of NetApp's innovations over the years? add

What are the innovations and services being offered to help customers leverage their AI initiatives and achieve data mobility? add

What initiatives are being implemented to support customers at different stages of their AI projects? add

bolt Powered by CUBE AI

Sandeep Singh, NetApp

search

Christophe Bertrand

>> Welcome back to NetApp INSIGHT 2025 in Las Vegas. I'm Christophe Bertrand, principal analyst with theCUBE Research. I'm very, very pleased to have a technology conversation with Sandeep Singh, who is the SVP and GM of the Enterprise Storage. Big day for you.

Sandeep Singh

>> It is a blockbuster day for not only just NetApp, but across all of our customers and partners. We just introduced the NetApp AFX and the AI Data Engine. And while we've been accelerating customers AI workloads across hundreds of customers for many years with NVIDIA, these two introductions just changed the game entirely.

Christophe Bertrand

>> Exactly. And we're talking about a data platform now that suddenly got my attention in the keynote. So from your standpoint, obviously you're very focused on the products, and make sure you have the right products and the right solutions in place for your customers. As we're literally navigating difficult times with finding ROI in AI and overinvesting, maybe at times, but maybe not. So, what are the type of requirements that you've observed from your customers? What are they expecting from you, and what do you think will make a big difference in the end?

Sandeep Singh

>> Look. First of all, AI just has this tremendous potential for every organization. And every organization is currently looking at how do they gain an AI advantage across the board, whether it's productivity gains, delivering net new experiences, or just finding net new overall business models or revenue streams. Fundamentally, AI, especially in the enterprise, is a data problem. Data is the fuel for AI. And as customers are looking at starting on their AI journey, the first challenge becomes what is the right infrastructure to support their AI initiatives? But very quickly, it's about how do I get my AI, or data AI-ready? And this is where customers need not only the performance and the scale at the infrastructure level, they need all of the enterprise-grade capabilities. This is about secure multi-tenancy so that they can have their overall AI centers of excellence, and feed multiple data teams or agentic AI in the future. This is about having this data mobility so that data, wherever it sits, can be made accessible to the GPUs without creating copies, and while preserving all of the security permissions. This is about having the ability to seamlessly have on-premises and cloud stitched together. AI is inherently a hybrid cloud workflow.

Christophe Bertrand

>> Right.

Sandeep Singh

>> There's a ton of innovation that's happening in the public cloud tied to AI, and yet customer's data is sitting on-prem, and you need to be able to combine those. And lastly, AI thrives on data, but data has to be fundamentally secure.

Christophe Bertrand

>> Right.

Sandeep Singh

>> So, you need all of the security that enterprises count on. That's what customers are asking us about, and this is why it's so exciting to have the AFX and AI Data Engine.

Christophe Bertrand

>> Right. And lots of challenges that you've covered here in a nutshell. But our research here, theCUBE Research shows that cyber resiliency is definitely in need of more work across the board as well as the need to protect nascent AI infrastructure, and yet leveraging AI itself in the process of management, governance, and data protection. So maybe, we can touch on the two sides of the coin in a second. So, you mentioned that data has to be good. That has to be trusted, that has to be resilient. So if we think about the big topics here, data management and data classification, lots of silos today, big challenge, cyber resiliency, data resilience, whatever term we want to use. I'm going to say even maybe bigger challenge because now, the attackers are AI-powered themselves. And then you have to take a look at the evolution of AI itself. From an infrastructure standpoint, as you pointed out, this is not your HPC workflows of the past. It's not just a bunch of databases. It's not just unstructured data or file systems. It's a very different animal. So, let's double-click on the AFX. And it makes me smile when I hear the term disaggregation. Because when I started in this business 25 years ago or so, there was a startup called NetApp that came up with that disaggregation concept. Now, very different world. No question about that, but walk us through this. And why is it so different and so critical that you are doing it this way?

Sandeep Singh

>> Look. Fundamentally over the years, NetApp in its 30-year history has had multiple innovation ways. As the technology has continued to evolve, NetApp has innovated on behalf of our customers. It began with the filer of just a simplified overall appliance. Then it shifted to virtualization was the key overall technology wave. And we partnered with VMware, and enabled for customers just to have a super seamless, scalable performant virtualization experience across the board. And then it became about cloud and hybrid cloud, and we partnered with our public cloud partners and created this notion of first-party overall cloud services, as well as enabling this whole notion of a data fabric with data mobility. AI is that net new wave across the board, and customers are looking for us to continue to innovate on their behalf, to enable them to leverage AI initiatives. And that's exactly, fundamentally, what we're doing. How are we doing it in multiple different ways? Helping customers just be able to seamlessly have the right infrastructure with all of the enterprise-grade capabilities, with hybrid multi-cloud capability, with all of the cyber security capabilities built in. That's just helping them get their data AI-ready and to accelerate their AI initiatives. I'll also mention that what's super exciting is that we recognize that different customers are at different stages of their AI initiatives. Some are in POC stage. Some are in production deployment stage. So, we are putting all of the value of AFX, AI Data Engine, and making it available as a service via NetApp Keystone for our customers. This way, whether they're in POC stage, production deployment, it's seamless. They can adopt, scale, and accelerate their AI initiatives, and they can seamlessly scale with NetApp Keystone. That's all on everything that we're building for customers to be able to help them with their AI initiatives. But as you touched upon, behind the scenes, we're obviously leveraging AI internally in our own products in terms of continuing to give them the most secure storage on the planet. This is where it comes into real-time ransomware detection. It's not about detecting hours to days later. It is about making detection in primary storage within seconds to minutes, and then being able to inform the necessary security teams and enable rapid recovery from that standpoint.

Christophe Bertrand

>> Right. Yeah, that's when it got my attention. Rapid recovery, the ability to test the automation. And I mean, those are exactly what end users are telling us they want to see as capabilities. And they're actually very much more likely to purchase a solution that does that. That's what they've also told us. So I think you're real spot on. So clearly, what we're saying is the evolution candidly from storage to, really, data services, if I were to summarize. Thinking about that, how does that enable an organization, whether a government organization, public product, doesn't matter, to be more agile? How do you see that playing out? Because you're really moving the ball now to a different conversation. You're talking to the business.

Sandeep Singh

>> Absolutely. So if you look at it, basically a lot of the enterprises are evolving to, basically, being that internal service providers. Or whether it's government entities at their scale, they're also operating as a service provider across the board. That first level of flexibility comes into enabling customers to build a unified data foundation, collapse silos across the different application workloads. What does that enable for customers? Be able to get a consistent set of management, consistent automation, so I don't have to retrain. I can overcome talent shortages and skills gaps. Tied to that, it's consistent data security model. Otherwise, with inconsistent data security models, your weakest link basically compromises you. You also get consistent overall operational recovery workflows. Last thing you want is in a disaster situation, is for the administrators to be fumbling around with the recovery workflows. Here, with a unified data foundation, it's that consistency that's built in. And then fundamentally, consistent experience across on-prem and cloud. Customers want to unlock that flexibility and the agility of being able to balance their application workloads, whether on-premises or on the cloud. And NetApp uniquely is enabling that for our customers. There's a level of simplicity at scale that is achieved with that, with this unmatched flexibility, both at the workload level as well as at data level. And then when you talked about data services, this is about just continuing to deliver what customers need to be able to have, the workload-integrated data management services. So when you think about mission-critical application workloads running on database applications and getting application-consistent overall backup copies of data that they can have and instantly be able to recover in an application-consistent manner, that's flexibility. But that's also safety and security for customers because it's bread and butter for that.

Christophe Bertrand

>> Right. I mean, AI will create more data. You have to protect the data. That's the source for AI. You have to protect the models. You have to protect the metadata. And you're going to have to protect the data from the agents talking to the other agents to demonstrate causality, et cetera, et cetera. So, yes.

Sandeep Singh

>> And if I can take that one step further... I mean, that's just a wonderful segue, and what George talked about in the keynote is when you expand that and you look at your entire dataset, and you transform it, you vectorized it, you can develop the graph databases as well. And ultimately, when you think about... And fast-forward to agentic AI, that metadata fabric layer becomes so critically important to apply that structure over unstructured data, and then have essentially a knowledge graph. That can be that interface point that where all of your data has been classified and curated, and guardrails are available, and it gives you access only to the right set of agents with the right context of the data, we see that as just a tremendous forward-looking opportunity.

Christophe Bertrand

>> And that's how you create trust in data. And for agentic AI, it will be key. Real quick, how do you see the future from a data infrastructure perspective?

Sandeep Singh

>> Look, the future is bright. Every organization, every customer is in the process of how do they continue to modernize their infrastructure? That modernization, certainly, AI is a major trigger point for customers, but so is cyber resilience. Virtualization strategies, customers are exploring. So there's many different areas that are leading to customers to say, "I don't need to just refresh. I need to be thoughtful. I need to build the right data foundation that helps me modernize at the same time." And having that right data foundation is where we see the opportunity with the NetApp data platform. Customers can onboard in terms of building a unified data foundation with them. They can leverage cloud services, or storage as a service. They can get cyber resilient. And that ultimately leads them to getting their data AI-ready.

Christophe Bertrand

>> Excellent. Sandeep, thank you so much. We've covered a lot of ground. Very exciting time. Congratulations on the announcements. You're going to be very busy for the next few quarters, I believe.

Sandeep Singh

>> Thank you, Christophe. This is just a fantastic opportunity. We're super excited. We're very much looking forward to continuing to work with our customers and our partners.

Christophe Bertrand

>> Great. Well, thank you so much. And to our viewers, thank you so much for being here with us at NetApp INSIGHT 2025. My name is Christophe Bertrand, principal analyst with theCUBE Research. And we'll be right back, so please stay tuned.