Fast Path to AI With Dell PowerEdge and NVIDIA
October 28, 2025 | 1:40 PM - 2:00 PM UTC
Varun Chhabra
SVP, Product Marketing Dell Technologies
Jason Schroedl
Director Product Marketing, Enterprise Platforms Google
Rob Strechay
Dir./Principal Analyst & Host theCUBE Research

AI adoption is accelerating, and enterprises need infrastructure that keeps pace. Dell Technologies and NVIDIA are working together to simplify how organizations deploy and scale AI across existing data centers. The latest Dell PowerEdge XE servers, equipped with NVIDIA RTX Pro 6000 Blackwell Server Edition GPUs, deliver adaptable performance for a range of workloads — from generative and agentic AI to digital twins and advanced analytics. These systems are designed to integrate easily into current environments, supporting air-cooled operation and established software stacks.

Through the Dell AI Factory with NVIDIA, enterprises can experiment, scale and refine AI initiatives without major infrastructure overhauls. It’s a pragmatic path to building modern AI factories that balance innovation, efficiency, and flexibility. Hear from industry experts as they discuss how the latest PowerEdge and RTX Pro innovations are helping enterprises accelerate their AI journey.

search
Rob Strechay

>> Hello and welcome to theCUBE's coverage of NVIDIA GTC D.C. The speed of change in AI is remarkable and partnerships are fueling that momentum. Dell AI Factory with NVIDIA is an excellent example of helping businesses everywhere turn next-gen possibilities into a reality. I'm excited to be joined today by Varun Chhabra, SVP of product marketing at Dell Technologies, and Jason Schroedl, who's the director of enterprise platforms, product marketing at NVIDIA. Welcome on board both of you.
Jason Schroedl

>> Thanks. Great to-
Varun Chhabra

>> Hi, Rob. Thanks for having us.
Rob Strechay

>> I mean, obviously, this is like the event of the year, to put it mildly, from an AI perspective. We're really excited to have both of you guys on. I think one of the things we want to dig into is now we're a couple of years into this next-gen of AI and things have been going really well, really rapidly and a lot of changing. Varun, Dell AI Factory with NVIDIA has been available now and been making waves for several years. What are customers asking for next? What's next on your roadmap there?
Varun Chhabra

>> Yeah, Rob, it's pretty incredible how much momentum Dell and NVIDIA have jointly had in this market. It was maybe 18 months ago that we announced the Dell AI Factory with NVIDIA on stage with Jensen at GTC last year. And in those 18 months, the momentum has been really, really remarkable. We've had over 200 new releases in those 18 months. So, the pace of joint technical innovation between our companies has been absolutely smoking. And then, it shows up in the market momentum as well. We have over 3,000 customers. Yes, that's not a typo. 3,000 customers in 18 months using the Dell AI Factory with NVIDIA. And that would never have happened if Dell and NVIDIA had not committed together to creating turnkey solutions, easy to adopt, really targeted at what our customers are looking for, and have the agility to be able to adjust and react to needs as they evolve. As we all know, the AI space is moving at a pace like no other. So, in the last 18 months, that's the reason why we've had over 200 releases because we're really looking very carefully all the time at how our customers are adopting these solutions. What are the roadblocks and the challenges they're encountering as they go on their AI journey, and how can we adjust them? That brings us to this point now, which is really about enterprise AI adoption, and the number one thing we hear from enterprises as we go and talk to them about scaling their AI efforts is that they want things to be simple and they want their AI infrastructure to be able to fit into their existing data center architectures. Doing a rip and replace, while a lot of organizations are looking at, that means that things are going to take longer to get ROI efforts. So, it's very important to combine the longer-term planning with the ability to be able to adopt the latest and the greatest when it comes to AI infrastructure, the latest and the greatest GPUs from NVIDIA and the servers from Dell. It's very important that they're able to do that in a way that is non-disruptive. And that's really what we're talking about here with the latest generation of the PowerEdge XE servers, the 4UXE7740 as well as the 7745. Adaptability and bringing the latest innovation for AI is key to what we're doing there with those servers. These servers, really, I think of them as the Swiss army knife for AI and GPU-accelerated workloads. They let customers scale up to eight PCIE-based GPUs per server, including the brand new NVIDIA RTX Pro 6000 Blackwell Server Edition. What we're trying to do here is really equip the servers with the latest and the greatest, but then to do it in a way that slots into what customers are doing today. These servers are air cooled, so you don't need to either retrofit or bring in liquid cooling. And as I'm sure Jason will cover, these represent the latest and the greatest of AI innovation, GPU acceleration from NVIDIA. So, customers not only get the best and the latest and the greatest, but they're able to do that in a form factor that allows them to take advantage of what they have already today versus having to do a rip and replace or a complete redesign of their data center infrastructure.
Rob Strechay

>> Yeah, I don't think you get to 3,000 customers if you had to rip out your data center, redo all the power and cooling and all of that. That is just not the way things work, to put it mildly. And 200 releases, that's a lot of advancement as well, and very impressive, to put it mildly, and great validation for the solution. But that kind of flexibility doesn't mean much without the right acceleration for an organization's application, so let's dive a little deeper here, Jason. And why don't you tell us about the new NVIDIA RTX Pro 6000 Blackwell Server Edition?
Jason Schroedl

>> Yeah, no happy to. So, I think first of all, thrilled about the momentum we're seeing with the Dell AI Factory with NVIDIA, this tremendous partnership we have with Dell to drive really enterprise AI at scale, and bring the capabilities that both Dell and NVIDIA bring together in solutions that can fit into existing data centers, and that's really what this is all about. I mean, we're seeing, I'd say, three fundamental trends in the industry today. The first is a shift from traditional computing to accelerated computing. And the second is what we're seeing is data centers becoming AI factories. And the third is the continual evolution of AI at just a phenomenal pace, as we're moving from generative AI to agentic AI, incorporating our reasoning and even moving to bringing AI into the physical world, with what we're seeing as physical AI and industrial AI. And the RTX Pro 6000 Blackwell Server Edition is at the heart of that, really enabling a wide range of different use cases for AI in existing enterprise data center environments as they build out these AI factories. So, we consider the Dell portfolio of servers that run the RTX Pro 6000 Blackwell Server Edition, RTX Pro servers, and these RTX Pro servers have just a phenomenal breadth of use cases that enterprises can run. So, everything from, again, agentic AI and generative AI, leveraging some of the capabilities we've built together with Dell and use cases for things like content generation, code generation, digital assistance, or for a wide range of different visual computing use cases. We're seeing adoption in, again, physical AI environments and manufacturing things like digital twins or for a wide range of different existing enterprise use cases like data analytics or virtual workstations. All of those use cases can run on the RTX Pro 6000 Blackwell Server Edition and these RTX Pro servers from Dell. It is really a game changer for enterprises as they're looking to build out new AI factories to be able to fit these servers into their existing data centers. They're designed to fit in an air-cooled environment, to run everything that IT knows how to run. So, it'll run your hypervisors, your Kubernetes platforms, all of the applications and tools that it knows how to deploy. Again, that's the beauty of the RTX Pro 6000 Blackwell Server Edition, bringing the power of Blackwell and tremendous performance, again, across this wide range of use cases. Compared to the previous edition, we're seeing, I think it's multiple times for the performance for Blackwell versus the prior generation Ada Lovelace with L40S, or even versus Hopper from a variety of different use cases in AI inference. So, bringing that power of Blackwell to the enterprise to help them deploy AI, starting for a handful of use cases and then scaling across their data center as they build out these new AI factories.
Rob Strechay

>> Yeah, and I love that you brought up the use cases because, to me, that's where the rubber meets the road for these organizations and why they're at GTC is really to understand what can they do. And as you noted, it's not just about the AI, it's all of the infrastructure, the Kubernetes and everything else that goes around that, and all of the different partnerships you bring to bear there helps drive that forward, especially with the ISVs and things. So, I had the pleasure of chatting with some of your joint customers. And I want you to, in your words though, dive in and help us explore some of those customer use cases. And can you give us some examples of what some organizations are doing with this technology?
Varun Chhabra

>> I think, Rob, the thing that strikes anyone who has conversations with enterprises about AI is that this is a secular movement across industries, across geographies. There's horizontal use cases that apply to many different industries, but then there's also a lot of innovation happening at the vertical layer where the vertical specific use cases as well. I think if you think about the use cases that pop up and the industries that pop up the most, it's modernization with manufacturing, using things like vision systems and being able to do fault detection and things like that early. Healthcare, really, really important. Enhancing patient care, faster diagnostics, better post-procedure care, more focused on personalized medication and personalized care. And then, supply chain is another big one. And then, of course, financial services is also big one with fraud detection, more enhanced and customized services. And energy, oil and gas use cases, where there's traditionally been a big role with HPC, of course, but now we're seeing more and more of that blend of AI and HPC being brought to the analytics that organizations in the energy space too. And then, of course, public sector and government is a big, big area where there's so much focus on efficiency and better citizen care and better citizen services. Really, sky's the limit across all of these industries. And Jason, maybe you can cover a few specific customer use cases that we're seeing?
Jason Schroedl

>> Yeah. No, I'd be happy to. So, we're seeing phenomenal success with the joint solution we have with Dell, the Dell AI Factory with NVIDIA, bringing together these servers with RTX Pro 6000 Blackwell Server Edition, together with Dell Storage, together with networking, including Spectrum-X networking, BlueField networking and software with NVIDIA enterprise software, ISV software, bringing together a complete solution together with Dell and Dell services to deliver a wide range of different use cases. So, with these joint customers, we're seeing success across multiple different industries. And I've spoken to these teams recently and there's just phenomenal benefits they're already realizing from these deployments, whether they're starting small and beginning with a handful of use cases or as they look to grow and expand and scale across multiple different use cases in their enterprise. So, I was just talking yesterday with a team, actually a company in the public sector, that organization that is building out an AI factory, leveraging these RTX Pro servers, to enable, again, a wide range of different use cases, but one of the first things they're looking to do is to bring the power of agentic AI and reasoning to their internal employee base for literally tens of thousands of users across their organization to bring the capability to leverage their internal knowledge bases, their internal secure data stores with the right data security and make that available with agentic AI to their employees so they can tap into that data, gain the knowledge, and through essentially a digital assistant they can make available to those employees. And that's just one of the initial use cases they're looking at, but they're also looking at bringing in a variety of different applications and use cases in part, using the flexibility of the RTX Pro 6000 Blackwell serve Edition platform to be able to bring in ISVs that have validated on that platform. For example, in code generation or knowledge search or in areas in this case out to the factory floor where there are using things like computer vision. They are using things like factory automation, even industrial robotics. So, a really wide range of use cases. This just was one customer that is in the aerospace and defense industry, again, in the public sector that I met with yesterday, but we're seeing this wide range of different use cases in the pharmaceutical industry. Met with the team just last week about some of their use cases in accelerating drug discovery, usually using computational molecular dynamics running on the RTX Pro 6000 Blackwell Server Edition to deliver phenomenal speed ups. Again, we're seeing just fantastic price performance benefits with this platform for, again, whether it's things like genomics and drug discovery in pharmaceuticals or another customer, a doctor recently in the manufacturing sector. That's again, building out things like digital twins and industrial robotics for factory floor automation. Those are just a couple of the use cases we're seeing across the board in a wide range of different industries.
Rob Strechay

>> Yeah. No, I think, again, I think that is so exciting because I think these use cases are real. And this also the fact that you don't have to refactor your entire data center to get into this, and you can start and grow. I think, which actually brings up a really critical point. It's very clear that the combination of Dell's flexible servers and NVIDIA's universal data center platform can fit many of those needs. But Varun, how should customers think about the right setup for their use case, such as going from a single department to a global rollout of that application that they're using it for?
Varun Chhabra

>> Yeah, I think, Rob, that's so critical for enterprises when we talk to them today, because it even starts with before a single department, right? You're talking about POCs that are generally small scale. You're trying it out with a pilot set of end users. You then can scale it out to a full department, and then from there, you go on to other use cases that are either in different lines of business or across the entire company. So, there's various stages to the scaling, starting from the infancy of envisioning a use case POC, all the way out to global rollout. And a key requirement in those things is the ability to be able to start small and then scale and scale and scale, without having to retool your applications and without having to re-architect your infrastructure. And the things we're talking about here, the XE7740 and the 7745 are built for that scale up. So, Jason talked about a bunch of different use cases. Whatever those use cases are, the servers are built on this universal data center platform that is built for multiple different use cases. So, customers have the peace of mind to know that not only they can try certain single use case led workloads or they can scale to a variety of different workloads that may require many different ISVs. So, that is one dimension of scale. The other dimension of scale is starting small for a few set of users and then going up. The solutions we're talking about here really help with that as well. So, as I mentioned, the 7740 and the 7745, they support up to eight RTX Pro 6000 Blackwell Server Edition GPUs that are running at full 600-watt performance. As Jason mentioned, this is an air-cooled powerhouse that fits right into your existing data centers. But even if you don't need to go to eight GPUs at first, you can actually start smaller. Whether it's smaller test beds or smaller environments or edge locations like a branch office or a distribution logistics facility, could be a retail store. We have smaller form factors there as well. We have the 2U PowerEdge 770 or the 7725 that give you incredible performance and efficiency with just one or two RTX Pro GPUs running at low power. So, if you think about the entire spectrum here that I've talked about, you want to deploy AI right at the edge, where your data is being created, or you want to deploy it at a facility, a smaller facility, or you want to scale it out in your data center to multiple hundreds and thousands of users, whether they're internal or external. The set of server lines that we're talking about here in the the PowerEdge XE server lineup, because it's built on that universal data center platform of the RTX Pro GPUs from NVIDIA, it allows customers to have that seamless expansion journey across use cases and the number of users.
Rob Strechay

>> Yeah, I can see that being critical flexibility for these organizations. I love how this technology is really meeting organizations where they're at and growing with them as they go. I mean, the Dell PowerEdge and NVIDIA RTX Pro 6000 Blackwell Server Edition offer, what I say is adaptable, high-performance platform, so that you don't have to choose between what you need today and what you're going to need tomorrow for AI inferencing or fine-tuning at scale. Varun, if organizations are ready to see what's possible with this, what do you think they should do next? How do they get in touch with the Dell team to go and try out a proof-of-concept or something like that?
Varun Chhabra

>> Yeah, there's a variety of options for customers that Dell and NVIDIA have jointly created for our joint customers. I think the first thing you can do is go talk to your Dell or NVIDIA salesperson to understand what's the quickest way to get going. We have a bunch of professional services capabilities and offers as well that help customers not only with the technology piece, which is deployment, planning, support, et cetera, et cetera, and scaling, but even well before that. So, if your organization needs help with even understanding, "Well, I have hundreds of different use cases that are coming out of the woodwork. Every single department that I have wants to do something with AI. How do I prioritize? How do I get stakeholders aligned?" We have consulting workshops that Dell Professional Services offers that have been built jointly within NVIDIA that help get customers started even way before day zero. "How do I plan for success? How do I make sure that my stakeholders are aligned?" If you have challenges with getting your data ready, that's another big bottleneck that we see all the time, which is, "How do I get a handle on my data estate? My data is distributed across different locations, across edge, private cloud data centers or public cloud data centers. How do I get a handle on all of that data? Which of my data is good? Which one is maybe not so good to feed into our models and workloads or applications?" We can help with that as well. And all of these professional services, I can't emphasize this enough, have not been built by Dell in a silo. These have been built jointly with Dell and NVIDIA, working together to make sure that these technologies are easy to deploy and easy to get started with. And then, as I said, it's not just about the technology. It's very, very important that customers think about processes and stakeholder alignment to be able to select the most impactful from an ROI perspective AI use cases, and we can help with that as well.
Rob Strechay

>> Yeah, I think that's great advice, especially that last part about figuring out the ROI before you go down that path, because I think then you understand the expectations and you guys are great partners at being able to do that. I really appreciate you both coming on. Thanks, Varun and Jason, for helping me dive deep into this at NVIDIA GTC D.C.
Varun Chhabra

>> Thank you so much for having us.
Jason Schroedl

>> Yeah, great to be here.
Rob Strechay

>> Thank you. And thank you everyone who joined us from NVIDIA GTC D.C. There's much more innovation ahead and we can't wait to share it with you. Stay tuned for more on theCUBE, the leader in analysis and news.
person_outline 6
WW
William W.
mb
michael b.
KS
Ken S.
CW
Chad W.
MT
Mellisa T.
Frank E.