Shared tool developed for quantum and supercomputer systems

Quantum computers are a key emerging technology, particularly for specific kinds of problems that require enormous computing power. However, integrating quantum systems into existing supercomputers poses a challenge.

Researchers at the Technical University of Munich (TUM) have developed a tool that combines quantum and supercomputers and enables them to interact seamlessly. This approach has been demonstrated experimentally in collaboration with a team at the Leibniz Supercomputing Center (LRZ).

Quantum computers operate using qubits, which, unlike classical bits, can exist in multiple states simultaneously through superposition. Additionally, qubits can be entangled, allowing for new computational paradigms that outperform classical systems for specific tasks. However, quantum computers are not universally applicable and are not intended to replace traditional high-performance computing (HPC). Instead, they are envisioned as a complementary accelerator within the HPC landscape.

Integrating quantum systems into HPC environments is complex due to differences in architecture, interfaces, and control mechanisms. “By developing the hybrid tool called sys-sage, we have addressed some of these challenges,” says Martin Schulz, Professor of Computer Architecture and Parallel Systems at TUM and member of the LRZ Board of Directors.

Bridging the gap: The sys-sage library

The sys-sage library was developed originally as a central interface for supercomputers. It collects and organizes dynamic and static information on a computer system’s architecture and topology and makes this information available for applications or other system components.

While the architecture describes the basic structure of a computer, the topology shows how the components are physically and logically connected. It can be seen as a map of the system, in a sense.

The expansion of the sys-sage library presented in the current study now permits a unified representation of the system topologies of both quantum and high-performance computers. This results in a hybrid structure that connects the two systems via a unified interface and makes it possible for them to be used together.

Sys-sage then informs other software components to facilitate their tasks towards better system optimization. For example, it supports the selection of whether a task should run on a quantum or classical system based on its computational characteristics, or the mapping of the problem to the best resources in the respective topologies.

“With this architecture, developed as part of the Munich Quantum Valley initiative and the Munich Quantum Software Stack (MQSS), we’re laying the groundwork for the productive and efficient use of quantum computers in supercomputing centers,” Schulz adds.