HPE World Quantum Day | Dieter Kranzlmüller, Leibniz Supercomputing Centre

Clips
More from HPE World Quantum Day

Dieter Kranzlmueller

Chair of the Board of Directors

Leibniz Supercomputing Centre

Dave Vellante

Co-Founder & Co-CEO

SiliconANGLE Media, Inc.

play_circle_outline Emphasis on technical independence and European sovereign quantum capability

play_circle_outline Euro-Q-Exa Joins EuroHPC: LRZ Deploys Fourth Quantum Computer Among Six European Systems

play_circle_outline Integrating Quantum Accelerators into HPC: Scheduler-Driven Hybrid Workflows for Performance and Cost-Based Decision Making

play_circle_outline Blue Lion Supercomputer: Application‑Tailored Many‑Accelerator Nodes and Quantum Advantage for Simulations, Optimization, and Logistics

Info
Transcript

Dieter Kranzlmüller, Leibniz Supercomputing Centre

Dieter Kranzlmueller

Chair of the Board of Directors Leibniz Supercomputing Centre

Dave Vellante

Co-Founder & Co-CEO SiliconANGLE Media, Inc.

HOST

In this interview from HPE World Quantum Day 2026, Dieter Kranzlmüller, chairman of the board of the Leibniz Supercomputing Centre and professor of computer science at LMU Munich, joins theCUBE Research's Dave Vellante to discuss how quantum computing is maturing from theoretical promise into a practical accelerator within high-performance computing infrastructure. Kranzlmüller traces LRZ's evolution from running Germany's earliest large-scale computers in 1962 into a full IT service provider for scientific research. He explains why Euro-Q-Exa — LRZ's fourth ... Read more

explore Keep Exploring

Why is technical independence in quantum technology important right now, particularly for Europe's supercomputing infrastructure? add

Why is Euro-Q-Exa being developed and how does it fit into Europe's broader quantum computing and supercomputing strategy? add

What factors determine whether a given job is executed on the supercomputer or on the quantum computer — e.g., performance, cost/economics, and suitability for the task? add

How is the Blue Lion supercomputer being designed, and what role do you see quantum computing (as an accelerator) playing for application workloads such as quantum-mechanics simulations and optimization/logistics problems? add

bolt Powered by CUBE AI

Dieter Kranzlmüller, Leibniz Supercomputing Centre

search

Dave Vellante

>> Welcome back to World Quantum Day made possible by Hewlett Packard Enterprise. My name is Dave Vellante and Dieter Kranzlmueller is with us. He's the chair of the board of directors for the Leibniz Supercomputing Centre, LRZ. Welcome. Welcome to the show, Dieter.

Dieter Kranzlmueller

>> Pleasure to be here. Thank you.

Dave Vellante

>> So tell us more about LRZ's mission. I know you're accelerating scientific research, like many if not all of the labs, but take us through your real mission.

Dieter Kranzlmueller

>> So LSZ was founded in 1962 in order to run big computers. So computers were built by ourselves in a sense, and they found out it's too big to run at a university. So we have developed over time to being an IT service provider for the universities for science in general, and that's what we still do today. We are running big supercomputers. We are providing services for science and research. And we say we are helping the digitalization of science and research in general.

Dave Vellante

>> Excellent. And of course, there's a lot of discussion. I was recently in Barcelona, so much discussion around sovereign computing and the intersection with high performance computing, and of course quantum. So I wonder if we could talk about what you're doing with quantum systems like Euro-Q-Exa into Europe's supercomputing infrastructure. Why is technical independence in quantum so important right now? It's such a theme around the world.

Dieter Kranzlmueller

>> Well, to tell the truth, I would turn this around. I think the important thing is technical independence. And quantum is a field where we can achieve that because it's such a new technology. It's just kind of coming up as being useful. And we can do that in a sense to also kind of help ourselves our positioning with the global developments that we see every day. And with quantum, we can really kind of bring something new to the table and do it independently in Europe, in Germany, without the need from anybody else on the planet.

Dave Vellante

>> So what exactly is Euro-Q-Exa? I know it's a part of a broader EuroHPC effort, but what does the system actually enable? What are the things that you're able to do that perhaps you weren't able to do a decade ago?

Dieter Kranzlmueller

>> So on a European level, Euro-Q-Exa is one of six quantum systems that are being established in Europe in different nationalities. And for us, Euro-Q-Exa is the fourth quantum computer that we are bringing to the table. The idea and the hope is that quantum technology gives us another jump forward in computational performance. The perspectives, the possibilities, the potential that quantum computing is delivering fulfills our growing need for computational performance. And that's why we are doing it. And this European effort is really kind of trying to make sure that we are not only focusing on one quantum technology, but instead a diverse set of quantum computing technologies, and that we have more or less everything in place to make this as part of Europeans supercomputing activities available to science research and industry.

Dave Vellante

>> Thank you for that. So I think there are a lot of misconceptions about quantum. A lot of people see it as this standalone thing. Jen-Hsun Huang makes a comment and all the quantum stocks drop and then he does a 180 and they all go up. But you've been integrating quantum into HPC environments for quite some time now. Discuss that integration. Set the record straight. Why is that so critical versus thinking about quantum as just something off in the corner?

Dieter Kranzlmueller

>> So the potential of quantum, and that stems mostly from the fact that quantum acceleration is exponential in a sense, it is great. But many of these things are still theory. So in theory, the quantum computer can do anything a regular computer can do. So there's no question about that. The problem is we want to get there faster. And for us, the solution is we're not using the quantum computer, as you said, as a standalone technology, but we are combining it with the most powerful technology that we have, which is supercomputing. So in our approach, the supercomputers will still be used for most of those things that it is excellent in, and that means many things that we are doing with supercomputers today. But in addition to that, we will be kind of utilizing quantum computers for those things where we already know today that the quantum computer is superior. So those applications that can really benefit from quantum will be done on quantum computers and will be using the supercomputers, the driving vehicle. This means in essence that we are sending jobs to the supercomputer and the system decides whether it's done on the supercomputer or whether it's kind of given to the quantum computer. And I think that's the important thing about why we do integration and why we leave we should combine both of these things together.

Dave Vellante

>> And that decision is based on a function presumably of performance and economics and the right fit for the right job? Is that right?

Dieter Kranzlmueller

>> Absolutely. Exactly. Yeah.

Dave Vellante

>> It sounds simple, but I'm sure it's pretty sophisticated at the backend. So help us understand what a hybrid HPC quantum workflow actually looks like in practice. You mentioned the supercomputer piece, the quantum piece. Are there GPUs in the middle doing error correction? Help us understand that.

Dieter Kranzlmueller

>> Well, our next supercomputer called Blue Lion is a system that is built in a traditional way in a sense with general purpose compute nodes, with many accelerators in there, and also with a focus on what our application users need. Yeah. I'm also a person that doesn't believe in the fastest supercomputer will be the best one, but instead the best supercomputer is those that helps the applications to get the best performance that they need for their particular workloads. And with quantum, bringing that to the field, quantum is more or less an accelerator, an accelerator that is specifically using the laws of quantum mechanics to compute faster on those application where it's useful. And to name just two examples, for instance, applications in quantum mechanics will benefit from a quantum computer because it uses the same principles. But even more, if we think about optimization, logistics, these kinds of applications, that's where quantum computer can also help with these exponential features to explore the state space and to give us kind of accelerated performance. Now, if we use the supercomputer and those parts which can be accelerated, execute them on a quantum computer, then together we can get the best performance of both worlds and accelerate the application in total.

Dave Vellante

>> Dieter, I've been an analyst for a long time and one of the events that has always been on my radar is the supercomputing show and it used to be a very niche show. Now it's like one of the most important events on the planet. Everybody's there. It's the hottest thing going. And you see innovations around material science and biology and energy and weather and medicine. So I want to talk about CLARA. Help us understand what that is, how it connects AI, supercomputing, and quantum to go after things like neurodegenerative disease and the like.

Dieter Kranzlmueller

>> Well, CLARA is a center for artificial intelligence and quantum computing in system brain research. And the idea here is exactly that it's not only the hardware that you need today to succeed, you need also to make sure that you have an ecosystem and you have to use this by kind of being attached to this ecosystem. And CLARA does exactly that. We use CLARA as a weekly to work with excellent experts in the field using these technologies, making sure that we're not only kind of putting the technologies on the table, but we are also helping the people to use these technologies. CLARA is exactly that particular environment, that ecosystem that we need for being efficient on these technologies and for kind of utilizing them for what they are needed. And our own role, LSZ's role in CLARA is exactly that. What can an IT service provider running supercomputers bring to the table? So for us, CLARA is also an extension of our capabilities to use the technologies that are in our center.

Dave Vellante

>> And I think increasingly the world is becoming aware of some of the innovations that come out of labs like yours. What kinds of breakthroughs are possible when you combine large scale data, AI, quantum, and systems like this?

Dieter Kranzlmueller

>> Well, there's manifold applications in that domain. I mean, if we look today around, this is what we are doing science for. So in order to kind of solve some of society's biggest problems, the systems that we are utilizing are exactly helping that for. I mean, we all remember COVIDs and the virus, these are things where these systems are being utilized and being used. We're also taking a look into the future. The same domain, for instance, we're walking to personalized medicine so that the medical practitioner not only gives you kind of drugs, medicine that worked in most of the cases, the statistical approach that we have today, but we want to model the user. We want to model the patient as a digital twin in the supercomputer and test how certain treatments work with that. And that is what we kind of want to do in that particular field. CLARA is working on brain issues, where can we kind of help supercomputers to improve the treatment with that particular... And that's exactly where we get all these technologies and where the everyday person will benefit of what we are doing with these things.

Dave Vellante

>> So I have a two-part question for you. LRZ, obviously it's an innovation hub. I'm curious as to how you collaborate with other labs, whether it's like Oak Ridge or Livermore, many others around the world, is the collaboration extensive? Is it minimal? Is it competitive? And either way, what have you learned about bringing these different technologies and disciplines together to solve these really hard world problems?

Dieter Kranzlmueller

>> So in order to use these computers for the best of society, you need the best brains, so to say. And these brains are really scattered around the world at different places. The labs that you mentioned in the US, they have all kind of wonderful experts. So from our point of view, and of course also somehow influenced by the global situation, from our world, collaboration is not a nice to have, it's something that we really need. We want to work with the best people in order to solve scientific questions, to solve scientific problems. And that's why we are collaborating. So it's not a give and take, so to say, but it's together we can do more in a sense. And that's why collaboration is so important, not only with the labs that you mentioned, but really with the people, the humans that are there, and they find ways on how we can use these technologies to really bring us forward all together.

Dave Vellante

>> So you're bringing together different disciplines, obviously, from these world experts. And what are some of the learnings over the last couple of years that bringing together those kind of disciplines and different technologies have shown you recently?

Dieter Kranzlmueller

>> So that's a very good question. One of the challenges you have if we work with interdisciplinary with experts from different fields is that each of the fields has its own language, so to say. We talk about the same things, but with different words in the different domains. Now, computing and the technology really gives us a basis that is the same for everybody. The supercomputer, the quantum computer is not different in one discipline or the other one. And so by working with these things, you're not only kind of learning the language of the other person, of the other expert, but you also kind of learn how you can bring your own expertise and together, again, do something more that was not possible before. And again, with CLARA, back in mind as one of the good examples, we have this interdisciplinarity. I'm working with experts from medical sciences. Yeah? So I'm 100% computer scientists, so to say. And it's really, for me, interesting to understand the language that a medical expert kind of brings to the table. And if we understand each other, we can really take those things that I bring to the table with those things that they bring to the table, and then again, make something that is bigger and really solves those grand challenges as we used to call these scientific problems.

Dave Vellante

>> I want to ask you a question for my commercial customers who they're drowning in AI, trying to apply AI these days, and they don't really have the capacity and the bandwidth, the cycles to be worried about quantum. At the same time, sure as algorithm is there and it's going to be able to break a lot of the standard encryption. And so we've been advising some of our commercial clients at the very least, it's coming and you need to pay attention to this. What would you tell some of your colleagues or friends in the commercial world? When should they start paying attention to quantum?

Dieter Kranzlmueller

>> Well, I think you're absolutely right. I mean, AI is the topic of the day, but quantum is not far behind. I mean, we see that quantum computers are growing in a sense. I mean, with Euro-Q-Exa, we have 54 qubits today, but we'll get 150 later this year, and we're walking into thousands of qubit next year, so to say. So that the point really is we are at the entrance already with quantum, and we hope that quantum can also deliver us more performance per buck for instance, so that we kind of address the problem that we have with AI today where lots of electrical power goes into the systems, and quantum also has promises for those areas. Now, I wouldn't say that this is kind of waiting just ahead of us. It's around the next corner, so to say, but the time to get engaged now with the systems is today, and we see users today. I mean, if I tell you about the medical example, yeah? Those are scientists working today with those things that will be in hospitals in 10, 15 years from now on, but that's the usual development. We do kind of tackle the hardest problems in order to kind of build the basis for what we are getting to into the future. And maybe in 15, 20 years from now, your medical practitioner will have an iPad or equivalent that is kind of using the powers of a quantum computer to deliver the personalized medicine you needed for your treatment then. And so that's the timescale. So I'd say that's what we see. There is huge interests already from some of the first runners in industry. And I think I wouldn't wait too long. We are almost there, so to say.

Dave Vellante

>> Yes. I was at RSAC, the big security conference recently, and the talk was that the attackers, they're exfiltrating encrypted data today with the hopes that they'll be able to decrypt when quantum comes along. So post-quantum cryptography is something that people need to be thinking about today. So thank you for that. My last question is, looking forward, what's next for LRZ? Where do you see the biggest opportunities over the next few years in HPC, AI, quantum? What are you guys working on that excites you and what should we be looking for?

Dieter Kranzlmueller

>> Well, true. For quantum, we're looking forward to those error correcting systems and to higher scalability, more qubits, more performance, so to say. But there's a number of other technologies which are sounding very interesting. If we talk AI, photonics is one of the things. We have a photonic chip already running in the data center, again, as a first mover in that area, so to say. And there is a couple of other technologies where I'm really eager to get my hands on. We are trying as a data center to really open those things up to the users. That's what we are doing with these technologies. And I believe we will see some lots of fascinating technologies coming up in the near future exactly for the reason to fulfill our needs for computational power.

Dave Vellante

>> Yeah. So exciting time. I mean, I'm no expert, but you're seeing all kinds of different approaches and a lot of experimentation and good R&D work going on. There's , there's gate-based approaches, there's all kinds of ways to attack the problem. As they say, there's many ways to skin a cat. Dieter, thank you so much for spending some time with us here and really appreciate your insights.

Dieter Kranzlmueller

>> It's my pleasure. Thank you for the good questions.

Dave Vellante

>> You bet. And you're watching theCUBE Research Coverage of HPE Quantum Day. Keep it right there for more deep dive content. Thanks for watching.