Menu

Six Five Connected with Diana Blass: Why Liquid Cooling Now Extends to SSDs

Six Five Connected with Diana Blass: Why Liquid Cooling Now Extends to SSDs

From robots to real-time video streaming 🤖 GTC 2025 showcased the magnitude of AI’s impact. But as Diana Blass discovered, this isn’t just about flashy applications; it’s about fundamentally rethinking the infrastructure that powers them. A key piece of that infrastructure? Storage. And companies like Dell and Solidigm are stepping up to meet the unprecedented demands of AI workloads.

Catch highlights from featured tech leaders Mary Elizabeth Sheehan, Director of Global Alliances at LTIMindtree, Ihab Tarazi, SVIP & CTO, AI, Compute & Networking at Dell Technologies, Keith Townsend, CTO Adviser at The Futurum Group, and Solidigm pros, Cody Noland, Thermal Mechanical Engineer, Avi Shetty, Senior Director of AI Market Enablement & Partnerships, Ty Macadam, Senior Sales Account Executive, alongside Jacob Yundt, Director of Compute Architecture at Coreweave. They emphasize that partnerships are critical for handling the resource-intensive requirements of AI workloads and enabling businesses to translate AI’s potential into tangible results.

Key takeaways include:

🔹Storage at the Forefront of AI: As AI applications become more sophisticated, they require massive datasets delivered at lightning speed. This is driving the development of advanced storage solutions like Solidigm’s liquid-cooled, fanless systems, which are crucial for optimizing performance and efficiency in AI-centric data centers.

🔹Addressing the Heat Challenge: The increasing power density of AI systems is pushing cooling demands to the limit. Solidigm’s innovative liquid-cooled SSDs directly address this challenge, enabling data centers to support more compute capacity.

🔹Powering the AI Explosion: Companies like CoreWeave, a major provider of cloud infrastructure for demanding AI projects, are relying on high-capacity storage solutions from Solidigm to manage the explosive data growth generated by multimodal LLMs and other AI applications.

🔹A Holistic Data Center Transformation: The AI revolution necessitates a comprehensive overhaul of the data center, where storage, compute, and networking must work in perfect harmony.

Learn more at Solidigm.

Watch the full video at Six Five Media, and be sure to subscribe to our YouTube channel, so you never miss an episode.

Or listen to the audio here:

Disclaimer: Six Five Connected with Diana Blass is for information and entertainment purposes only. Over the course of this webcast, we may talk about companies that are publicly traded, and we may even reference that fact and their equity share price, but please do not take anything that we say as a recommendation about what you should do with your investment dollars. We are not investment advisors, and we ask that you do not treat us as such.

Transcript:

Diana Blass: Welcome to the AI Super Bowl, also known as Nvidia GTC. This is a huge conference in the AI industry. A chance to see what’s to come in the market, but also a chance to tackle some of the big questions.

Jensen Huang: How do you solve the data problem? How do you train the model? How do you scale?

Diana Blass: Let’s go check it out.

Diana Blass: Inside the show floor, AI isn’t just code in the cloud anymore.

Ihab Tarazi: Joy is our favorite dog.

Diana Blass: This is the rise of physical AI.

Mary Elizabeth Sheehan: She is responding in real time.

Diana Blass: A new era that goes beyond generating an answer. Instead reasons, plans and acts.

Jensen Huang: It understands things like friction and inertia, cause and effect.

Diana Blass: And when physical AI is harnessed inside a machine, you get human innovation like this.

Nicholas Ringborg: Hi Emma, how can I pay my water bill online?

Diana Blass: Meet Emma, a digital assistant deployed on the City of Amarillo’s government website.

Nicholas Ringborg: So you can go on amarillo.gov and actually interact with Emma and get information about the City of Amarillo and some government services.

Diana Blass: Emma is a product of Dell and Nvidia’s AI factory, which enterprises lean on to deploy AI applications without having to make costly investments of hardware, software and skilled talent.

Ihab Tarazi: What makes it unique is that they get the combination of everything Dell makes and everything Nvidia makes, completely packaged and ready to go.

Diana Blass: It’s a glimpse into the future. As Jensen predicts.

Jensen Huang: Every company that has factories will have two factories in the future. The factory for what they build and the factory for mathematics. The factory for the AI.

Diana Blass: But making that a reality, not so easy.

Nicholas Ringborg: It’s a completely complex space where you need to think about the infrastructure itself, the AIOps layer, the ML OPS layer, and how all of that stack fits together with the compute, the storage, the networking and all of that.

Diana Blass: It’s why data centers are being reimagined. We’ve already seen GPUs and CPUs move into liquid cooling systems. But would you be surprised to learn that storage is heading in that direction too?

Keith Townsend: Yeah, if you would have asked me a couple of years ago, I would have told you that’s a gimmick.

Diana Blass: From gimmick to necessity. You see, storage has become a critical piece of making AI work. It’s responsible for holding massive data sets and moving them to GPUs at lightning speed.

Diana Blass: But as data centers become increasingly packed with power hungry systems, storage has come under pressure too.

Keith Townsend: Without a doubt, when SSDs get hot, the performance slows. So if you need sustained performance, liquid cooling, SSDs make sense.

Diana Blass: It’s why we were excited to see this solution at GTC, a liquid cooled fanless storage system unveiled by Solidigm.

Cody Noland: Where we’re taking a cold plate and we’re directly attaching it to a single side of our SSD and dissipating the heat.

Avi Shetty: So what typically happens, Diana, is you have liquid, the coolant kind of coming in and touching a cold plate. The cold plate, this slots into a chassis. The cold plate touches the SSD and then the SSD internally cools both sides and the liquid kind of flows out.

Diana Blass: This solution is innovative on multiple levels. Not only does it drive more power efficiency into the data center, but it also reclaims physical space by removing the fans. That means more space for compute and engineers can service the unit more easily thanks to the integration of hot swapping.

Cody Noland: We can remove the SSD, service it if needed. We can gain access to the backplane if needed.

Diana Blass: This system is the latest way that storage systems are modernizing for the AI age. And of course, that extends to the drive itself.

Avi Shetty: Out of the box, I have a 122 terabyte.

Diana Blass: Enter Solidigm’s 122 terabyte drive. A drive that packs massive capacity into a smaller footprint.

Ty Macadam: We help reduce the power using larger capacity, less drives or less servers, and overall TCO savings from power and cooling.

Diana Blass: For customers like CoreWeave, it’s a game changer. CoreWeave provides the cloud infrastructure behind today’s most demanding AI projects.

Jacob Yundt: We’ve seen explosive growth in multimodal LLMs. So like this text, video, audio, image generation is just generating and consuming an incredible amount of data. Like every time we think we’ve understood what our forecast is, we get it wrong and we just get like a bajillion more drives, bajillion more servers. And basically none of that would be possible without these high cap drives because.

Ty Macadam: He said he’s doing bajillion servers.

Jacob Yundt: Bajillion. Bajillion servers. That’s the official term.

Diana Blass: You heard them. Bajillion. Skyrocketing data that’s placing new demands on infrastructure. At GTC, we saw how vendors like Solidigm are stepping up to meet those challenges. But storage is just one part of the puzzle. Stay tuned as we continue to unpack the tech powering the AI era. Signing off for Six Five Media. I’m Diana Blass.

Author Information

Diana Blass

Diana Blass is a journalist with a background in technology news and analysis. Her work has appeared on Fox Television Stations, The Discovery Channel, CRN, Light Reading, and other Informa-owned media brands. In addition to her work at The Six Five, she manages Diana Blass Productions, where she develops and produces digital documentaries, podcasts, and commercials for media and corporate brands.

Related Insights
Cohere’s Multilingual & Sovereign AI Moat Ahead of a 2026 IPO
February 20, 2026

Cohere’s Multilingual & Sovereign AI Moat Ahead of a 2026 IPO

Nick Patience, AI Platforms Practice Lead at Futurum, breaks down the impact of Cohere's Tiny Aya and Rerank 4 launches. Explore how these efficient models and the new Model Vault...
Will NVIDIA’s Meta Deal Ignite a CPU Supercycle
February 20, 2026

Will NVIDIA’s Meta Deal Ignite a CPU Supercycle?

Brendan Burke, Research Director at Futurum, analyzes NVIDIA and Meta's expanded partnership, deploying standalone Grace and Vera CPUs at hyperscale, signaling that agentic AI workloads are creating a new discrete...
Will Salesforce’s Latest Acquisition Provide Momentum For its Agentic Workflows
February 20, 2026

Will Salesforce’s Latest Acquisition Provide Momentum For its Agentic Workflows?

Keith Kirkpatrick, VP and Research Director at Futurum, covers Salesforce’s acquisition of Momentum, and discusses the need to clearly lay out the vision and timing for integrating the technology into...
CoreWeave ARENA is AI Production Readiness Redefined
February 17, 2026

CoreWeave ARENA is AI Production Readiness Redefined

Alastair Cooke, Research Director, Cloud and Data Center at Futurum, shares his insights on the announcement of CoreWeave ARENA, a tool for customers to identify costs and operational processes for...
Arista Networks Q4 FY 2025 Revenue Beat on AI Ethernet Momentum
February 16, 2026

Arista Networks Q4 FY 2025: Revenue Beat on AI Ethernet Momentum

Futurum Research analyzes Arista’s Q4 FY 2025 results, highlighting AI Ethernet adoption across model builders and cloud titans, growing DCI/7800 spine roles, AMD-driven open networking wins, and a Q1 guide...
Cisco Live EMEA 2026 Can a Networking Giant Become an AI Platform Company
February 16, 2026

Cisco Live EMEA 2026: Can a Networking Giant Become an AI Platform Company?

Nick Patience, AI Platforms Practice Lead at Futurum, shares insights direct from Cisco Live EMEA 2026 on Cisco’s ambitious pivot from networking vendor to full-stack AI platform company, and where...

Book a Demo

Newsletter Sign-up Form

Get important insights straight to your inbox, receive first looks at eBooks, exclusive event invitations, custom content, and more. We promise not to spam you or sell your name to anyone. You can always unsubscribe at any time.

All fields are required






Thank you, we received your request, a member of our team will be in contact with you.