The International Conference for High Performance Computing, Networking, Storage, and Analysis
The International Conference for High Performance Computing, Networking, Storage, and Analysis
Authors: Bob Sorensen (Hyperion Research), Yonatan Cohen (Quantum Machines), Mathias Pütz (ParTec AG), Santiago Núñez-Corrales (National Center for Supercomputing Applications (NCSA)), James Palles-Dimmock (Quantum Motion Technologies Ltd), Kristel Michielsen (Forschungzentrum Juelich), Ugo Varetto (Pawsey Supercomputing Centre)
Abstract: Integrating quantum computing (QC) test beds into scientific computing environments presents challenges in software interfaces and system familiarity. High-performance computing (HPC) centers are adopting this task but selecting suitable test bed technologies is complex due to numerous providers with varying maturity levels and the associated risk of single vendor systems.
A component-based approach is promising but faces challenges with the lack of standardized benchmarks, and the need for device-specific calibrations. This discussion addresses the challenge of component-based approaches and explores unifying access to diverse QC technologies, leveraging HPC for optimization, and fulfilling researcher needs.
Long Description: Many HPC centers are now tasked with incorporating quantum computing (QC) test beds into their scientific computing environments. This poses challenges for HPC centers as early QCs appear alien not only in the machine room but also in their software interfaces, programming models and usage. However, HPC centers have a successful history of adopting new technologies, making them a natural choice for this task. Moreover, classical HPC and QC demonstrate synergies, as only specific parts of larger programs can be off-loaded to the QC, while the remaining problem components reside on the HPC system. Additionally, the computation pipeline of the QC often relies on significant classical computing resources, suggesting an inherently hybrid future for HPC and QC.
Selecting the right technologies for quantum test beds is challenging, as hundreds of different technology providers offer solutions in different stages of maturity. While some, including IBM, Google or DWAVE, aspire to become full stack providers, many others also specialize in addressing one or two specific layers required for implementing a so-called full stack. Integrating relatively mature cloud offerings from Google or IBM into existing HPC environments also proves challenging as co-scheduling of local HPC and QC resources is not trivial. Determining which quantum computer will drive the anticipated advances and assessing the risks associated with investing in a single technology or vendor Are critical considerations. Although research and development of benchmarks to gauge the performance of hardware-specific resources are in early stages, the lack of robust quantum benchmarks further amplifies the risks.
Ideally, such early test beds would greatly benefit from the ability to seamlessly switch qubit technology with minimal disruption, enabling exploration of alternative paths or upgrades within the same family. Adopting a best of breed component-based approach, similar to the prevalent modular HPC environments, for different layers in the full stack has significant appeal. This approach offers decision makers enhanced flexibility, investment protection, and potential benefits for key research areas already in active development such as error correction code landscape understanding, hardware-independent compilation and programming frameworks, and quantum benchmarking. However, this promising component-based approach currently faces obstacles such as lack of standards, the need for device-specific calibrations, algorithmic optimization, and expertise of skilled integrators capable of harmonizing the components into a cohesive “whole”.
In the BoF, we will present concepts in addressing the challenge of component-based approaches. We will explore strategies for unifying access to diverse QC technologies and leveraging HPC to optimize QC systems. We aim to stimulate an engaging discussion among the audience and panelists regarding how this approach can effectively address their specific needs. Furthermore, we aim to identify the most significant obstacles that must be overcome in order to successfully implement these concepts.
Back to Birds of a Feather Archive Listing