Hugues Meyrath (LinkedIn) has spent more than three decades in enterprise infrastructure, watching the storage industry evolve from spinning disks to cloud-native object storage and now AI-driven data management.
Before becoming President and CEO of Quantum in 2025, he held executive leadership roles at Dell Technologies, Dell Technologies Capital, EMC, Brocade, Juniper Networks and ServiceChannel. His relationship with Quantum stretches back even further: early in his career he worked on components for Quantum hard drives and DLT tape technology, later joined the company as an employee, competed against it during his years at Dell EMC, served on Quantum’s Board of Directors, and ultimately returned to lead the company.
Digital Production: You’ve worked for Quantum, competed against Quantum, served on its Board of Directors and are now back as President and CEO. That’s a rather unusual career arc. What convinced you to come back?
Hugues Meyrath: It honestly feels personal. My first job out of university involved manufacturing read and write heads for Quantum hard drives. Later we were building heads for Quantum’s DLT tape drives, long before LTO became the standard everyone knows today. So Quantum has been part of my professional life almost from day one. Like many people in technology, I moved around over the years. I worked inside Quantum, later in investment banking when the company split into two businesses, and eventually I found myself at Dell EMC, where Quantum suddenly became one of our competitors.
Looking back, I realized I’d seen the company from almost every possible perspective. When I joined the Board a few years ago it already felt like coming home. Becoming CEO afterwards wasn’t something I had mapped out twenty years earlier, but sometimes careers have a funny way of making a complete circle.
DP: Digital Production actually goes back to the ADIC days. We still remember receiving exactly one invoice from ADIC before suddenly everything carried a Quantum logo.
Meyrath: (Laughs.) That’s probably one of the shortest reseller transitions in history.
DP: Out of that acquisition came StorNext, which is still one of the industry’s most widely used media file systems. More than twenty years later, what makes you proudest about it?
Meyrath: Honestly? The fact that customers are still running it every day. That’s rewarding because it tells you the product solved a real problem instead of chasing a trend. Media companies don’t buy storage because storage is exciting. They buy it because editors expect footage to appear instantly, colourists don’t want dropped frames and rendering farms need predictable throughput.
StorNext has always been designed around that idea. It’s less about absolute benchmark numbers and more about guaranteeing that hundreds of users can work simultaneously without getting in each other’s way. Another thing that has become increasingly important is intelligent tiering. People often assume every piece of data belongs on flash. That’s usually not true. Some projects are actively being edited. Others might not be touched for weeks but still need to remain immediately searchable. Some have to be retained for years.
Moving data automatically between flash, disk and tape lets customers balance performance, cost and energy consumption without constantly making manual decisions. That idea has become much more valuable today than it was even five years ago.
DP: At the European partner conference earlier this year you gave a keynote that felt remarkably optimistic. Walking around the Quantum booth here at NAB, that optimism seems genuine. The atmosphere feels different.
Meyrath: I hope people notice that. When I became CEO, one of the first things I wanted to change was our relationship with partners. Companies like Quantum don’t succeed alone. We never have. Partners build businesses around our technology. They support customers locally. They solve deployment problems. Very often they’re the face of Quantum long before anyone ever speaks to us directly.
I felt we’d lost sight of that. So rebuilding trust with the partner ecosystem became a priority from day one. The response has exceeded my expectations. We’ve had record bookings, record backlog and record revenue over the past quarters. That’s obviously great financially, but what really matters is what it says about confidence. Our partners decided Quantum was worth betting on again.
DP: Walking around your booth this week, customers seem genuinely pleased to see Quantum becoming more visible again.
Meyrath: That’s actually one of the nicest surprises. Customers walk over and tell us they want Quantum to succeed. As a CEO you obviously like hearing that, but more importantly it reminds you that enterprise technology is ultimately a relationship business. People buy products, but they also buy confidence. They want to know you’ll still be there in five or ten years. That’s exactly the trust we’re rebuilding.
Power, Not Capacity: Why Tape Is Suddenly Interesting Again
DP: One topic kept coming up in your presentations this week: energy. A few years ago every storage discussion revolved around performance or capacity. Now everyone seems to be talking about electricity. What changed?
Hugues Meyrath: AI changed the conversation. People naturally focus on GPUs because that’s the visible part of the AI revolution. They’re impressive machines, they’re expensive, and they’re what everybody likes to photograph.
But once you look underneath the GPUs, you quickly discover a much bigger challenge. Where does all the power come from? That’s becoming one of the defining questions for the entire industry. You can already see it happening around the world. There are regions where new data centres simply can’t be approved because the electrical infrastructure isn’t there.
Once you start looking at storage through that lens, everything changes. Storage isn’t just about buying the fastest system anymore. It’s about asking yourself how much electricity you’re willing to spend keeping data online. I often joke that our largest Scalar tape library here at the booth consumes less power than my gaming PC, yet it can store well over one hundred petabytes of uncompressed data. People laugh when I say that, but then they stop for a second and realize the comparison is actually true. Electricity is becoming one of the most valuable resources inside the modern data center.
DP: That’s not the sales pitch tape enjoyed ten years ago.
Meyrath: (Laughs.) No. Back then we mostly talked about backup. Today we’re talking about infrastructure. That’s a very different conversation.
DP: Tape is suddenly fashionable again. That’s a sentence I genuinely didn’t expect to write in 2026.
Meyrath: I don’t think many people did. The funny thing is that tape itself hasn’t changed as dramatically as the software around it has. Most people still imagine the old workflow. You had a backup application writing cartridges sequentially, somebody took them off-site, and hopefully you never had to look at them again.
That’s the picture many people still have in their heads. Modern tape environments look nothing like that. Today you have flash handling immediate workloads, traditional disk taking care of active datasets, and tape acting as the long-term storage tier. The software continuously moves information between those layers based on policies, usage and cost. The user doesn’t think about cartridges anymore. They think about their data.
DP: So tape has effectively disappeared behind the software.
Meyrath: Exactly. That’s really the important shift. If I ask someone where their files are stored in Amazon S3, they usually can’t answer. They don’t know which disk, which rack or which building their object lives in. They simply ask for the object, and the system retrieves it. That’s the experience we’re bringing into private infrastructure.
The software exposes a modern object interface, while underneath the data moves intelligently between flash, disk and tape. The user doesn’t have to understand the mechanics. They simply get the performance they need while the infrastructure optimizes cost and energy consumption in the background. Ironically, tape has become more valuable because people no longer have to think about tape.
Storage technology works best when nobody notices it. If editors, artists or researchers spend their day thinking about storage, something has probably gone wrong. The infrastructure should quietly make intelligent decisions for them.
DP: One misconception probably comes from scale. Most people still picture a single tape drive sitting somewhere underneath a desk.
Meyrath: (Laughs.) I know exactly the picture you’re talking about. People remember swapping cartridges by hand. What we’re building today has very little in common with that. Take our Scalar i7 library as an example. It can hold thousands of cartridges, dozens of tape drives operating in parallel, robotic automation, object storage interfaces and policy-driven data management. You’re not manually loading media anymore. The library simply becomes another storage resource inside your infrastructure. People often think tape is slow. Randomly searching through tapes manually would be slow. Searching indexed objects while dozens of drives work simultaneously is an entirely different experience. That’s why software matters so much.
DP: During your keynote, you mentioned object storage several times. That’s a term people hear frequently, but not everyone understands what it means.
Meyrath: It’s actually a very simple idea. Traditional file systems organize information into folders and directories. Object storage treats every piece of information as an object with its own metadata. That makes it much easier to scale because you’re no longer relying on a traditional directory hierarchy.
It’s also the model cloud providers popularized through services like Amazon S3, which has become something of a common language across the industry. By supporting those interfaces, customers can build workflows that behave very much like cloud infrastructure while keeping the data inside their own facilities. For media companies that’s becoming increasingly attractive because they want cloud-like flexibility without necessarily moving hundreds of petabytes into public cloud storage.
DP: And those petabytes are growing rather quickly.
Meyrath: Faster than most people appreciate. AI generates enormous amounts of data. Camera resolutions continue to increase. Scientific datasets become larger every year. Companies rarely delete information anymore because storage is often cheaper than recreating knowledge. The result is that almost everyone is trying to solve the same puzzle. How do you keep more data while consuming less power? That’s exactly why tiered storage has become so relevant again.
DP: Is that why hyperscalers suddenly started buying tape again?
Meyrath: It’s one of the reasons. If you look at overall tape shipments, the market has changed dramatically. Roughly speaking, last year around fifty exabytes were written to tape. This year we’re looking at something closer to one hundred and forty exabytes.
That’s growth. You don’t see numbers like that because people suddenly became nostalgic. You see them because economics changed. Power matters. Total cost of ownership matters. Long-term data retention matters. Tape solves those problems extremely well when it’s integrated into a modern storage architecture.
DP: Which also means tape has quietly evolved from a backup medium into active infrastructure.
Meyrath: That’s the best way to describe it. Backup is still important, of course. But today tape is becoming part of intelligent data management. It’s another tier in a workflow rather than the destination at the very end. Once customers make that mental shift, they start looking at storage very differently.
AI Agents, Unified Storage and Why CEOs Should Walk the Show Floor
DP: You can’t walk ten metres at NAB this year without someone mentioning AI. Every company has an AI strategy, an AI assistant or an AI roadmap. Looking ahead a couple of years, do you think more AI agents will eventually be interacting with Quantum storage than human users?
Hugues Meyrath: I actually think that’s quite possible. People naturally associate AI with chatbots because that’s what they see every day. But infrastructure is a very different world. Think about what happens inside a large storage environment. You might have billions of files, millions of objects and thousands of users spread across multiple locations. Someone has to decide which data belongs on flash, which can move to disk, which should be archived to tape, what needs another copy, what suddenly becomes important again, or whether unusual access patterns might actually indicate a security issue. Those aren’t glamorous decisions, but they’re exactly the kind of repetitive, data-driven tasks AI is exceptionally good at.
People sometimes ask whether AI will replace storage administrators. I don’t really look at it that way. I think AI will help them manage environments that have become far too large and far too complex for any individual to optimize manually. It’s less about replacing people and more about letting them focus on the decisions that actually require experience and judgement.
DP: So AI becomes another layer inside the storage system itself?
Meyrath: Exactly. We’re already using AI internally in software development, just like many companies are. But what excites me much more is what happens after deployment. Imagine storage continuously observing how data is used, learning access patterns, identifying unusual behaviour, recommending better placement or automatically optimizing where information lives. Storage stops being passive. It becomes something that actively helps manage itself. I think we’re only scratching the surface of what’s possible there.
DP: Looking ahead two years, what is on Quantum’s roadmap?
Meyrath: Bringing everything together. Historically we have built excellent technologies, but like many companies that have grown over decades, those technologies evolved in parallel. We have flash products, disk products, tape products and data management software. Each has its own history, engineering teams and architecture.
What we’re doing now is rethinking that stack. The customer shouldn’t have to think about which product they’re interacting with. They should think about their data. If the platform automatically chooses the right storage layer, moves information and presents a consistent experience, then we’ve done our job.
DP: In other words, flash, disk and tape stop being separate products and become different “personalities” of the same storage system.
Meyrath: That’s a nice way of putting it. Customers don’t buy storage because they enjoy comparing interfaces or hardware generations. They buy it because they need reliable access to their information. The infrastructure should adapt to the workload, not force customers to redesign their workflows every time new hardware appears. That’s really the philosophy behind where we’re heading.
DP: One thing that stood out during our conversation is that you’re talking surprisingly little about hardware. Most storage vendors traditionally lead with specifications.
Meyrath: Hardware will always matter. You still need fast flash, reliable disks, high-capacity tape libraries and robust networking. But hardware alone isn’t enough anymore. The value comes from orchestration. How do all those technologies work together? How do they become easier to operate instead of more complicated every year? I think that’s where the industry is heading. Customers are asking for simpler infrastructure, not more infrastructure.
DP: That’s a refreshing answer from someone running a hardware company.
Meyrath: (Laughs.) Maybe that’s because I’ve spent enough years in this industry to know customers rarely wake up in the morning thinking, “I wish I had another storage appliance.” They want their projects delivered on time. If we can help them achieve that while reducing operational complexity, then we’ve created real value.
Twenty-Five Years of NAB
DP: Last question. NAB has only just opened. What are you hoping to take home from this year’s show?
Meyrath: Feedback. I’ve probably been coming to Las Vegas for twenty-five years. I used to live in California, so this became almost a second office. Between trade shows, customer events and the occasional birthday celebration, I’ve spent more time here than I’d probably like to admit. At some point, you’ve seen enough casinos.
What never gets old is the people. You walk through these halls and suddenly you’re looking at twenty-five years of your own career. Former colleagues, customers, partners, competitors. People you’ve worked with, people you’ve competed against, people you’ve solved impossible problems with. That’s one of the wonderful things about this industry. Companies change, job titles change, business cards change, but the relationships stay remarkably constant.
Of course we compete. That’s part of business. But there’s also a genuine sense of community. I regularly meet CEOs from companies we compete with. We shake hands, catch up, talk about the industry. Many of us have worked together at some point during the last twenty or thirty years. Storage is a relatively small world. Media and entertainment is similar. You build relationships over decades, and that’s something I appreciate.
DP: Finally, after everything we’ve talked about today, from AI to energy, what do you hope people remember about Quantum over the next few years?
Meyrath: That we’re making storage simpler. We often think infrastructure becomes more complicated as technology advances. I believe the opposite should happen. The technology underneath can become incredibly sophisticated. Flash, disk, tape, cloud, AI, automation, policy engines, object storage, cybersecurity. Customers shouldn’t have to worry about any of that. They should simply know their data is safe, cost-effective and ready whenever they need it.
This interview was conducted by Jürgen Firsching during NAB 2026 in Las Vegas and has been edited for length and clarity.
