Analysis revealed earlier this month in the science journal Nature used NVIDIA-powered supercomputers to validate a pathway toward the commercialization of quantum computing.
The analysis, led by Nobel laureate Giorgio Parisi, focuses on quantum annealing, a technique that may someday resolve complicated optimization issues that are difficult for traditional computer systems.
To conduct their analysis, the crew utilized 2 million GPU computing hours on the Leonardo facility (Cineca, in Bologna, Italy), almost 160,000 GPU computing hours on the Meluxina-GPU cluster in Luxembourg, and 10,000 GPU hours from the Spanish Supercomputing Community. Moreover, they accessed the Dariah cluster in Lecce, Italy.
They used these state-of-the-art assets to simulate the habits of a sure form of quantum computing system often known as a quantum annealer.
Quantum computer systems essentially rethink how info is computed to allow solely new options.
In contrast to classical computer systems, which course of info in binary — 0s and 1s — quantum computer systems use quantum bits or qubits that may enable info to be processed in solely new methods.
Quantum annealers are a particular kind of quantum laptop that, although not universally helpful, might benefit from fixing specific optimization issues.
The paper, “The Quantum Transition of the Two-Dimensional Ising Spin Glass,” represents a significant step in understanding the part transition — a change within the properties of a quantum system — of Ising spin glass, a disordered magnetic material in a two-dimensional airplane, an essential drawback in computational physics.
The paper addresses the issue of how the properties of magnetic particles organized in a two-dimensional airplane can abruptly change their habits.
The research additionally reveals how GPU-powered methods play a crucial function in creating approaches to quantum computing.
GPU-accelerated simulations enable researchers to grasp the habits of complicated methods in creating quantum computer systems, illuminating probably the most promising paths ahead.
Quantum annealers, just like the methods developed by the pioneering quantum computing firm D-Wave, function by methodically reducing a magnetic area that’s utilized to a set of magnetically vulnerable particles.
When strong enough, the utilized area will act to align the magnetic orientation of the particles—just like iron filings will uniformly stand to consideration close to a bar magnet.
If the sector’s power is assorted slowly enough, the magnetic particles will organize themselves to reduce the vitality of the ultimate association.
Discovering this secure, minimum-energy state is essential in a very complicated and disordered magnetic system often known as a spin glass since quantum annealers can encode sure sorts of issues into the spin glass’s minimum-energy configuration.
Discovering the secure association of the spin glass then solves the issue.
Understanding these methods helps scientists develop higher algorithms for fixing challenging issues by mimicking how NNature offers complexity and dysfunction.
That’s essential for advancing quantum annealing and its functions in fixing extraordinarily tough computational issues that presently don’t have an identified environment-friendly resolution—issues that are pervasive in fields ranging from logistics to cryptography.
In contrast to gate-model quantum computer systems, which function by using a sequence of quantum gates, quantum annealers enable a quantum system to evolve freely in time.
This isn’t a common laptop—a tool able to perform any computation given adequate time and resources—but it might have benefits for fixing explicit units of optimization issues in utility areas such as car routing, portfolio optimization, and protein folding.
By conducting in-depth simulations on NVIDIA GPUs, the researchers realized how key parameters of the spin glasses making up quantum annealers change throughout their operation. This permits a greater understanding of how to use these methods to realize a quantum speedup on vital issues.
A lot of the work for this groundbreaking paper was first offered at NVIDIA’s GTC 2024 know-how convention. Learn the complete paper and study more about NVIDIA’s quantum computing work.