Nvidia today announced its Quantum Optimized Device Architecture (QODA), a new computing platform that aims to bridge the gap between quantum and classical applications. QODA marks a break from Nvidia’s generally proprietary position and relies on an open approach to quantum computing and its integration with classical systems similar to those we are used to managing in our daily lives.
NVIDIA’s QODA aims to make it easier for IT professionals to leverage the strengths of quantum computing (QC) in application development and workload acceleration techniques. QC removes the zeros and ones from the classical bit and replaces them with qubits, a unit that is now approaching mainstream and can actually represent both states at the same time, opening up vast speed-ups in previously infeasible problems.
With the number of different approaches to quantum computing, from IBM’s own superconducting transmon qubits, to silicon quantum dots, topological superconductors, trapped ions and other technologies, the current quantum computing landscape is not too different from that of the first semiconductors. days. But QODA’s high-level language will support all types of quantum computers and its compiler will be available as open source software.
Image 1 of 2
Nvidia sees QODA as the glue that ties it all together. Acting as a software bridge, Nvidia’s QODA will allow developers to choose to run their quantum tasks on GPUs or quantum processors. All that separates them is a simple function call.
Quantum computers and their hardware processor, the quantum processing unit (QPU), are ideally suited for simulating processes at the atomic level. This could enable fundamental advances in chemistry and materials science, triggering domino effects in everything from more efficient batteries to more effective drugs.
In separate experiments, researchers from the Jülich Supercomputing Center near Cologne used quantum machine learning to simulate how proteins bind to DNA strands and classify satellite images of Lyon, France.
Tricky logistical issues – such as the traveling salesman problem, deliveries, passenger pickups or even flight routes – are also solved for maximum efficiency through quantum computing. Gone are the days of logistics solutions such as UPS “never turn left” navigation system.
In one experiment, a quantum computer recently installed in Jülich showed the most efficient way to deliver nearly 500 flights, demonstrating the technology’s potential in a way that will directly impact people’s lives. Life is already short, why waste it on delayed planes?
While several quantum computers, from embedded systems to scalable and individual QPUSs, are already available, none have yet achieved the speed, reliability, ease of deployment, and ease of use to tackle relevant business tasks. But researchers see a way forward: hybridization.
“For a long time, we’ve had a vision of hybrid systems as the only way to achieve practical quantum computing – tied to today’s classical HPC systems, quantum computers will give us the best of both worlds,” said Kristel Michielsen, who leads the quantum program at the Jülich Supercomputing Center near Cologne. “We can’t go on with today’s mainstream computers alone because they consume so much power and they can’t solve some problems,” she continued. “But paired with quantum computers that won’t consume as much power, I think there could be the potential to solve some of our most complex problems.”
HPC and AI specialists will be able to take advantage of Nvidia’s QODA in a familiarly classic environment, easily able to take advantage of the probabilistic approach to computing that is the hallmark of quantum.
“Scientific breakthroughs can occur in the short term with hybrid solutions combining classical computing and quantum computing”, said Tim Costa, HPC and Quantum Computing Product Manager at Nvidia. “QODA will revolutionize quantum computing by providing developers with a powerful and productive programming model.”
Nvidia’s QODA provides access not only to today’s quantum processors, which are actually simulated on Nvidia’s SuperPod GDX (opens in a new tab) GPU-based systems and accelerators, but also those of tomorrow. QODA builds on Nvidia’s work in cuQuantum, a highly specialized software application that allows Nvidia customers to develop individual quantum circuits, simulating their performance and characteristics before actual deployment.
With QODA, developers can build complete quantum applications simulated with NVIDIA cuQuantum on GPU-accelerated supercomputers, opening doors that have so far remained closed.
To top off its QODA announcement, Nvidia at the Q2B conference in Tokyo announced QODA collaborations with today’s leading quantum hardware vendors: IQM Quantum Computers, Pasqal, Quantinuum, Quantum Brilliance and its diamond-based qubits, and Xanadu with its impressive Borealis QPU. Quantum QC Ware Software Suppliers and Computer Zapata have also partnered with Nvidia and supercomputing centers Forschungszentrum Jülich (opens in a new tab)Lawrence Berkeley National Laboratory (opens in a new tab)and Oak Ridge National Laboratory all threw down their gloves with Nvidia’s solution.
That doesn’t mean Nvidia’s Approach will be the exclusive platform for those Approaches, or anything like that. But it demonstrates confidence in the company’s solution as another piece of the quantum computing puzzle. And what a glorious puzzle it is.