Ion-Trap Quantum Computer Ready for Novel Research and Development at the LRZ

Industry / Press Release September 17, 2024

September 16, 2024 -- The next quantum computer is ready for use in research and development at the Leibniz Supercomputing Centre (LRZ) of the Bavarian Academy of Sciences and Humanities. The system is based on trapped-ion technology and computes with 20 qubits. It was funded by the Bavarian Ministry of Sciences and the Arts (StMWK) and the Ministry of Economic Affairs, Regional Development and Energy (StMWi) with around €9.8 million as part of the Hightech Agenda Bavaria.

Ion-Trap Quantum Computer Ready for Novel Research and Development at the LRZ
Working on the new ion-trap quantum computer: Hossam Ahmed, engineer from the LRZ quantum team, prepares the AQT system for the operating parameters in the LRZ Quantum Integration Centre (QIC). Photo: MQV | Jan Greune

Installing innovative computing technology in a data center environment typically takes about a year before it is deemed ready for reliable operation. However, the LRZ team from the Quantum Computing and Technology (QCT) department, in collaboration with experts from AQT, successfully completed the installation ahead of this timeline. Although some adjustments were necessary to meet the operational parameters of the computer room – such as dust filtering and vibration – the system is now fully operational and performing computations. Starting in October, researchers from MQV and its partner organizations will be able to access and utilize the system as part of an early user program.

HPC acceleration with various quantum systems

The AQT quantum computer is a promising technology that the LRZ is making available to the scientific community for computing and processing research data. The system consists of an ion trap, laser, camera unit and control electronics. It operates on the basis of electrically charged atoms (ions) whose quantum states are controlled by laser beams. A laser can address arbitrary pairs of the 20 qubits in the quantum register as required. This full connectivity, along with lower error rates, allows it to achieve similar or better results compared to systems with a higher qubit count but less connectivity. Furthermore, the AQT system requires no extensive infrastructure for cooling or power supply.

The LRZ plans to integrate the ion trap quantum computer as an accelerator into its supercomputer environment. The QCT team is already experimenting with classical processors and components from its own testbed at the Quantum Integration Centre (QIC) to develop the necessary interfaces. The LRZ has already successfully integrated a quantum computer earlier this year into the LRZ supercomputer SuperMUC-NG based on superconducting circuits as part of its hybrid quantum-HPC development efforts.

To efficiently operate these hybrid systems, researchers at the MQV are working on system software and tools for quantum computers and on coordinating the computation between various quantum accelerators and traditional supercomputers. Their work includes creating hybrid algorithms tailored for end-to-end calculations and identifying critical interface points between the systems to establish an allover standardization.

A single software stack for different systems

The goal and vision of all these preparations is to operate quantum technologies holistically and to accelerate supercomputers using various quantum processors. Ideally, users will have access to multiple quantum processors and will be able to use those that best suits the needs of their research application. The realization of this endeavor, however, will only be possible with the right software to support the integration of various quantum technologies into high-performance computers. This software, known as the Munich Quantum Software Stack (MQSS), is being developed within MQV in close collaboration between LRZ and TUM. It includes not only monitoring and control tools for innovative quantum and hybrid computer systems, but also optimized programming models that enable scientists to code their applications in a familiar manner without requiring extensive knowledge of the quantum architecture. The AQT device, with its computation driven by laser-trapped ions, serves as a critical target for the development of the MQSS .