Collaborative Research: PPoSS: Planning: Software Stack for Scalable Heterogeneous NISQ Cluster

点击翻译按钮获取中文摘要

Development of large-scale and practical quantum computers is a priority for many countries, industries, and researchers. Demonstrating quantum computers at scale will change the computing model as it is currently known forever, enabling scientific discoveries at an unprecedented pace. This project’s novelties are in designing future quantum systems as a cluster of heterogeneous quantum computers. Such an approach is significantly different from all existing endeavors, as it will be cost effective, scalable, more usable, and more reliable. The project’s impacts include outlining the challenges in such systems, proposing solutions, engaging the community, and describing a plan to build a full software stack for such heterogeneous quantum-computing-based clusters. The project will also engage the multidisciplinary quantum computing community through three invited workshops to inform the potential path towards solutions for the challenges outlined. Through a backbone stakeholder committee, the project will ensure sustainable and sustained workforce development and broadening participation in computing objectives, outcomes, and impact at scale. In addition, the project personnel have a strong commitment to increasing participation of underrepresented groups (including women, racial minorities, and persons with disabilities) in planned activities.This project explores the feasibility of designing a full software stack for a cluster of heterogeneous Noisy Intermediate-Scale Quantum (NISQ) machines. The project will make contributions to the: (a) Realization of cluster of heterogeneous NISQ machines as a quantum-computing platform with large-scale simulation and evaluation on a real platform; (b) Programming environment and user interface to provide a visual interface to understand quantum noise; (c) Compilation techniques to account for heterogeneity of NISQ machines and temporal errors; (d) Runtime to enable fault-tolerance, resource management and scheduling considering the queuing time and noise condition in real time with the help of a resource monitoring mechanism to query the calibration information from all available quantum computers; (e) Co-design of the stack with quantum machine learning and quantum chemistry applications; (f) Utilization of the system calibration data from the multiple existing quantum machines, then apply fidelity degradation detection on each noise attributes to generate the fidelity degradation matrix which is used to define multiple new evaluation metrics to compare the fidelity between the qubit topology of the quantum machines; and (g) Engagement of the multidisciplinary quantum computing community through three invited workshops to inform the potential path towards solutions for the challenges outlined. Education, workforce development (WFD) and broadening participation in computing (BPC) are a major priority of this project. These will be realized as: (a) Through a backbone stakeholder committee, the investigators will ensure sustainable and sustained WFD and BPC objectives, outcomes, and impact at scale. The project plan capitalizes on the breadth of expertise of the PIs with an overall strategy organized to reach increasingly larger stakeholder groups (starting from project members, the broader systems community, and finally to K-12 and non-affiliated professionals); (b) In addition, the project personnel have a strong commitment to increasing participation of underrepresented groups (including women, racial minorities, and persons with disabilities) in planned activities; (c) The investigators will incorporate research outcomes in multiple courses; and (d) The project will facilitate collaboration and synergy among systems researchers, and engage and partner with industry for technology transfer and commercialization.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.