Universal Benchmark for Assessing the Performance of Quantum Devices

Scientists from the Department of Energy’s Oak Ridge National Laboratory have come up with a quantum chemistry simulation benchmark that helps assess the performance of quantum devices and steer applications development for next-generation quantum computers.

The study outcomes have been reported in npj Quantum Information.

In quantum computers, the laws of quantum mechanics and units called qubits are used to considerably improve the threshold at which transmission and processing of information can be performed. In contrast to conventional “bits” that have a value of 0 or 1, qubits are encoded with both 0 and 1, or any combination thereof, which facilitates a huge number of possibilities for data storage.

Although quantum systems are still only in their infancy, they have the ability to be exponentially more robust compared to existing superior classical computing systems. The systems also hold the potential to completely transform research in chemistry, materials, high-energy physics, and across the wider scientific spectrum.

However, since these systems are in their early stages, gaining insights about the applications that may be best suited to their exclusive architectures is regarded as a crucial field of study.

We are currently running fairly simple scientific problems that represent the sort of problems we believe these systems will help us to solve in the future. These benchmarks give us an idea of how future quantum systems will perform when tackling similar, though exponentially more complex, simulations.

Raphael Pooser, Principal Investigator, Quantum Testbed Pathfinder Project, ORNL

Pooser and his team calculated the bound state energy of alkali hydride molecules on 16-qubit Rigetti Aspen and 20-qubit IBM Tokyo processors. Since these molecules are simple and their energies are well perceived, the quantum computer’s performance can be effectively tested by the researchers.

The researchers tuned the quantum computer as a function of a few parameters, thereby calculating the bound states of these molecules with chemical accuracy, which was achieved through simulations on a classical computer. The fact that systematic error mitigation was also part of the quantum calculations is equally important, elucidating the flaws in existing quantum hardware.

Systematic error appears when the “noise” intrinsic to existing quantum architectures has an impact on their functioning. Quantum computers are highly delicate (for example, the qubits that the ORNL researchers use are stored in a dilution refrigerator at about 20 mK (or over −450 °F), thus temperatures and vibrations from the environments around then can lead to instabilities that affect their accuracy.

Such noise might render a qubit, for instance, to rotate 21° in the place of the preferred 20°, which significantly affects the results of a calculation.

This new benchmark characterizes the ‘mixed state,’ or how the environment and machine interact, very well. This work is a critical step toward a universal benchmark to measure the performance of quantum computers, much like the LINPACK metric is used to judge the fastest classical computers in the world.

Raphael Pooser, Principal Investigator, Quantum Testbed Pathfinder Project, ORNL

Although the calculations were quite simple than what is feasible on superior classical systems like ORNL’s Summit, which is presently the most powerful computer in the world, quantum chemistry, together with quantum field theory and nuclear physics, is regarded as a quantum “killer app.”

Simply put, it is considered that as quantum computers evolve further, they will have the capability to carry out a broad array of chemistry-related calculations in a more efficient and more accurate manner, better than any classical computer that is currently being used, including Summit.

The current benchmark is a first step towards a comprehensive suite of benchmarks and metrics that govern the performance of quantum processors for different science domains. We expect it to evolve with time as the quantum computing hardware improves. ORNL’s vast expertise in domain sciences, computer science and high-performance computing make it the perfect venue for the creation of this benchmark suite.

Jacek Jakowski, Quantum Chemist, ORNL

For over a decade, ORNL has been planning for revolutionizing platforms like quantum through dedicated research programs in quantum materials, quantum computing, sensing, and networking.

The ultimate goal of these efforts is to expedite gaining insights into what ways near-term quantum computing resources can be useful in dealing with the current most daunting scientific challenges and supporting the recently launched National Quantum Initiative, which is a federal effort to ensure American leadership in quantum sciences, specifically computing.

Leadership such as that will require systems such as Summit to ensure the steady progress from devices like those used by the ORNL researchers to larger-scale quantum systems that are exponentially more robust than any existing systems in operation.

The Quantum Computing User Program at the Oak Ridge Leadership Computing Facility offered access to the Rigetti and IBM processors. The facility also offers early access to today’s commercial quantum computing systems and supports the development of future quantum programmers by offering educational outreach and internship programs.

This study was supported by the DOE’s Office of Science Advanced Scientific Computing Research program.

“This project helps DOE better understand what will work and what won’t work as they forge ahead in their mission to realize the potential of quantum computing in solving today’s biggest science and national security challenges,” added Pooser.

The next step for the researchers would be to calculate the exponentially more complex excited states of these molecules, which will enable them to develop more innovative error mitigation schemes and bring practical quantum computing possibilities a step closer to reality.