SEQUOIA End-to-End

Transparent quantum software engineering and algorithm design application-centric end-to-end solutions
Logo SEQUOIA End-to-End

The SEQUOIA End-to-End project worked to develop transparent, automated, and controllable end-to-end solutions for the industrial use of hybrid quantum applications and algorithms through holistic quantum software engineering.

Quantum computing offers great potential to advance classical computation running on current high-performance computing (HPC) infrastructures to a new level. Although quantum computing is still in its infancy, the arrival of the first quantum computers (QC), such as the IBM Q System One, now enables academia and industry to begin developing exciting solutions that are not feasible on current computer systems.

The project SEQUOIA End-to-End enabled researchers and industry stakeholders to evaluate the opportunities that quantum computing offers, and to test them on real-world scenarios from a multitude of application domains, such as manufacturing, robotics, energy, mobility, and the health sector, to name but a few. SEQUOIA End-to-End focused on three key areas:

Transparency and further development of the quantum software development process

SEQUOIA End-to-End tackled current challenges that were identified in the current development process, starting with the identification of an industrial quantum computing use case, continuing with the selection of an optimal QC algorithm, and ending with the execution on a quantum computer and the evaluation and interpretation of results. Here, problem-specific algorithm design for hybrid applications was key.

Quantum software engineering

It is of utmost importance to develop and employ easy-to-use tools, toolchains, approaches, best practices and systematic design throughout the development process.

End-to-end solutions

The project implemented eight industrial use cases that were identified in the areas of production, logistics, engineering, and automotive. Demonstrators and application-specific benchmarks were developed from these and in order to transfer knowledge to industry and society.

The essential goal of SEQUOIA End-to-End was to make the bottlenecks in the entire quantum software development process transparent and to research and provide performant, automated, and controllable end-to-end solutions through holistic quantum software engineering. The results will enable the industrial use of hybrid quantum applications and algorithms in the future.

HLRS’s contributions to SEQUOIA End-to-End aim to pave the way for hybrid solutions that benefit from the integration of HPC, AI, and quantum computing.

Project achievements

HLRS is strongly interested in how quantum processors will eventually fit into the compute continuum. We envision that in the medium to long term, quantum processors with a compact form factor and low latency will be available within a high-performance cluster. During the SEQUOIA End-to-End project, we:

  • Explored novel hybrid algorithms for quantum computing involving classical-quantum co-processing.
  • Benchmarked the performance of classical emulators on widely used QC algorithms, which is essential for research within the QC field and for calibrating real quantum devices.
  • Examined the performance of current quantum computing systems in comparison with classical computing architectures.
  • Explored the benefits of quantum machine learning (QML) for use cases that involve solving partial differential equations (PDEs).

Future objectives

  • Based on our early explorations of QML, we wish to more comprehensively benchmark its potential advantages for realistic datasets and quantum circuit sizes.
  • As a continuation of our benchmarking activities, we plan to explore quantum optimization problems in more depth; e.g., schemes for encoding the quantum state. We would like to compare these best quantum approaches to state-of-the-art classical solutions for the same problem; e.g., via CPLEX and Gurobi.
  • Extending our work with solving PDEs we will explore methods for solving more complex PDEs involving the use of quantum gradients and potentially time evolution.

Project partners

  • Fraunhofer-Institut für Arbeitswirtschaft und Organisation IAO (Koordinator)
  • Fraunhofer-Institut für Angewandte Festkörperphysik IAF
  • Fraunhofer-Institut für Kurzzeitdynamik Ernst-Mach-Institut EMI
  • Fraunhofer-Institut für Produktionstechnik und Automatisierung IPA
  • Eberhard Karls Universität Tübingen, Lehrstuhl Eingebettete Systeme
  • FZI Forschungszentrum Informatik
  • Universität Stuttgart, Institut für Höchstleistungsrechnen HLRS
  • Albert-Ludwigs-Universität Freiburg, Lehrstuhl für Theoretische Physik
  • Karlsruher Institut für Technologie, Institut für Informationssicherheit und Verlässlichkeit

Predecessor project SEQUOIA

If you would like to learn more about the predecessor project SEQUOIA, which
was funded by the WM Baden-Württemberg from 2021 to 2024, we
refer you to the project page: www.iaf.fraunhofer.de/en/researchers/quantum-systems/quantumcomputing/sequoia.html

Funding

Contact