Details of HPE quantum computing are emerging

Quantum computing, artificial intelligence and high-performing computing will play complementary roles in driving the next era of technological innovation. That is, if the industry can find a way to open the doors and enable the work to be developed more easily than the quantum experts.

Speaking to CUBE business editor, Paul Gillin (pictured, left), as part of the AnalystANGLE segment during the event HPE World Quantum Day event, Dave Vellante (right), senior analyst at CUBE Research, explained that quantum, HPC and AI are not locked-in competitors in a zero-sum game of technological dominance.

“It’s really about combining these three powerful technologies — CPUs, GPUs and quantum processors, or QPUs — to solve previously unsolvable problems,” Vellante said.

One remaining obstacle is that the industry is stuck in the “Stone Age” when it comes to quantum computing software development, according to Gillin. He emphasized the need for “Python for quantum” which would simplify and democratize access for developers to quantum systems.

“Until that can be made more widely available — until more people have access to quantum technology — it’s going to continue to be kind of a lab project,” Gillin said.

The latest HPE World Quantum Day eventduring a special broadcast on CUBE, SiliconANGLE Media’s live streaming studio, Gillin and Vellante spoke with industry experts from Oak Ridge National Lab, Argonne National Laboratory and more about how quantum computing, HPC and AI are starting to converge – not as competing technologies, but as part of the next generation of architecture. (* Disclosure below.)

Here’s part of the analysis with Gillin and Vellante, part of SiliconANGLE and CUBE covering HPE World Quantum Day event:

Here are three important details you might not miss from HPE World Quantum Day:

Insight #1: The short-term impact of Quantum will be on speeding up existing workflows.

Researchers are exploring ways in which quantum computing can improve and accelerate critical science that is already being done with the help of supercomputers.

Rather than replacing older systems, quantum computing is likely to enter business as a highly targeted accelerator – designed to improve specific, complex workloads where traditional methods fail. Tom Beck (pictured)head of the scientific interaction division and leader of the quantum-HPC acting group at Oak Ridge National Laboratory – home of Frontier, one of the world’s most powerful supercomputers – described this interim phase in the conversation with Vellante, positing the quantum not as a replacement, but as a complementary force within hybrid computing environments.

“We’re trying to figure out how to connect quantum computers to HPC so we can offload certain parts of the math to the HPC machines to speed those up, but make a very difficult quantum part or something that can be quantum accelerated in a quantum device,” Beck said. “So it’s really a game of information transfer. How do you speed up the flow of information between two machines and do something very difficult in a quantum device?”

Here’s the full CUBE interview with Tom Beck:

A team at Argonne National Laboratory is similarly working to integrate quantum computing into real-world applications, which could help. speed up processing times application of cases in chemistry and materials science, according to Laura Schulzproject leader for quantum innovation.

“When we try to simulate or study quantum mechanical effects with HPC, we have to build a simulation,” Schulz told CUBE. “Quantum computing allows researchers to study quantum mechanical effects” directly.

This supports a balance of work where quantum mechanics is relied upon to solve part of the problem and its results are returned to ancient equations running on supercomputers for every task.

Here’s CUBE’s full interview with Laura Schulz:

Insight #2: Researchers are working to solve engineering challenges that prevent mainstream discovery.

Quantum provides a significant improvement in processing power over supercomputing, making it particularly useful for complex problems that require scale, such as tracking the behavior of neutrinos in stars.

Commercial applications in areas such as logistics and drug research are already within reach — modern lasers and MRI machines rely on quantum science — but physical and engineering barriers continue to hinder widespread adoption, according to Kristi Beckdirector of the Livermore Center for Quantum Science at Lawrence Livermore National Laboratory.

“Applications in pharmacology come from the anticipated benefits we have of many chemical problems that, ultimately, support our understanding of drug interactions,” Beki told CUBE. “But I would say that those are too far because of the difficulty of those issues, even though they are the ones where we can expect to see a better win than what we get in planning.”

Here’s the full CUBE interview with Kristi Beck:

Back at Oak Ridge National Laboratory, researchers are working with quantum vendors, universities and labs to understand how it’s done. make quantum more accessible. This means examining requirements at all levels of the technology stack to create a middleware layer that can handle the unique complexity of quantum mechanics, according to Amir ShehataHPC systems engineer, Quantum-HPC Group, Oak Ridge National Laboratory. Qubits, the fundamental unit of information in quantum computing, present a particularly thorny challenge.

“When you’re working with superconducting, qubits have short life cycles and degrade quickly, so you have a tight time frame to deal with,” Shehata explained. “And when you’re working with neutral atoms or some other kind of slow motion, there are different timing constraints there. So your software stack that you’re developing has to handle all these kinds of demands that are thrown at you from the hardware side.”

The new software infrastructure to enable quantum computing will essentially rely on mainstream technology, such as GPUs, Shehata added.

Here is CUBE’s full interview with Amir Shehata:

Insight #3: Increasing the power of quantum computing will mean solving the right task at the right time.

Since qubits can exist in many states at once and are connected by entanglement, quantum computers can simultaneously consider all possible solutions to a problem. This opens up opportunities to effectively solve challenges that would take today’s supercomputers years to figure out.

Quantum computers are best suited for complex mathematical calculations that require high precision, according to Mikael Johanssonmanager of quantum technologies at CSC, the Finnish IT Center for Science.

“Just take the green revolution as an example; we can create better catalysts and develop next-generation batteries and super-central magnets, of course, for today’s society,” Johansson explained.

Here’s CUBE’s full interview with Mikael Johansson:

But that doesn’t mean quantum computers will replace supercomputers, he said Dieter Kranzlmüllerchairman of the board at the Leibniz Supercomputing Center. They both have a job to do.

“This means, in effect, that we send tasks to a supercomputer and the program decides whether it is done by the supercomputer or whether it is given to a quantum computer,” Kranzlmüller told CUBE. “I think that’s an important part of why we’re doing the merger and why we believe we should bring these two things together.”

Here’s CUBE’s full interview with Dieter Kranzlmüller:

https://www.youtube.com/watch?v=waNoc6NpNq0

However, progress is contingent broad participation. The Pawsey Supercomputing Research Center in Perth, Australia, is developing a program called Setonix-Q. Its purpose is to provide Australian researchers with the tools and guidance to explore quantum mechanics, according to him Pascal Elahiquantum supercomputing research lead at Pawsey.

“We want to expand access not just to quantum computing researchers, but to people who want to solve the problem,” Elahi told CUBE.

Here is CUBE’s full interview with Pascal Elahi:

Find our complete HPE World Quantum Day issue:

https://www.youtube.com/watch?v=videoseries

(* Disclosure: TheCUBE is a paid media partner of HPE’s World Quantum Day event. Among HPE, the CUBE’s event sponsor, or other sponsors have editorial control over content on The CUBE or SiliconANGLE.)

Photo: SiliconANGLE