ERI: Harnessing Quantum-Classical Computing with a Cloud-Edge Framework for Cyber-Physical Systems

网络物理系统（CPS）和量子计算的云边缘计算已经独立发展，各自应对独特的挑战和机遇。Cloud-Edge-CPS研究主要集中在使用经典资源，对量子设备可能提供的潜在好处的探索有限。这在理解量子技术如何增强Cloud-Edge-CPS性能和功能方面留下了知识空白。与此同时，量子计算社区主要专注于开发高量子比特、稳定的硬件和完善底层技术，而对拓宽低量子比特量子机器的应用场景关注不够，这在噪声中尺度量子（NISQ）时代更容易获得。该项目旨在通过将量子边缘节点和量子云等配备量子的设备纳入云边缘协作计算框架，弥合这些研究领域之间的差距。它强调了低量子比特量子机器在资源受限环境中的潜力，并将其引入新的使用场景。此外，它还促进了各种研究和开发社区之间的合作，包括云边缘计算，网络物理系统和量子计算。拟议的量子经典系统强调可扩展性，为来自这些社区的研究人员和工程师创造一个支持性的环境，最终刺激这些学科的创新与合作。该项目将开发一个配备量子的Cloud-Edge协作计算框架，以有效管理资源受限设备上的异构参与者、网络通道和量子噪声。所提出的框架由三个主要组件组成：（i）位于远程数据中心的云服务器，提供充足的量子和经典资源;（ii）位于终端设备附近的边缘节点，其资源比云少，可以分为两种类型-量子经典边缘和仅经典边缘;以及（iii）充当系统消费者的终端设备，其拥有最少的资源并且可能配备有量子处理器。基于该框架，它提供了各种模块，包括终端设备注册，资源管理，任务建模和卸载估计，探索量子特性（如叠加，纠缠和隐形传态）的潜在优势。具体地，系统为每个终端设备构建客户端简档。当一个计算任务到达时，它会对一个特定的任务进行建模，并生成两个执行计划，量子经典的和仅经典的。基于预定义的量子服务，这些服务有可能为终端设备提供显著的好处（例如，二次或指数加速），系统根据计划预测任务执行时间，并选择满足客户端约束并最大化整体系统性能的计划。由于对量子机器的访问有限，该项目将实现分布式量子经典Cloud-Edge-CPS模拟器，以进行大规模的基于云的实验，并支持高度的异构性。此外，它还将开发一个运行时采样器，以研究量子噪声对资源受限的NISQ机器的影响，详细说明云如何帮助管理固有的噪声和错误。该项目的最终目的是通过整合量子设备并研究其在各种应用场景中的潜在优势，提高Cloud-Edge协作计算系统的性能、效率和能力。该奖项反映了NSF的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

Cloud-Edge computing for Cyber-Physical Systems (CPS) and Quantum Computing have evolved independently, each addressing unique challenges and opportunities. Cloud-Edge-CPS research has predominantly focused on using classical resources, with limited exploration of the potential benefits that quantum-equipped devices may offer. This leaves a knowledge gap in understanding how quantum technologies could enhance Cloud-Edge-CPS performance and capabilities. Simultaneously, the quantum computing community has primarily focused on developing high-qubit, stable hardware and refining the underlying technology, with insufficient attention to broadening the application scenarios for low-qubit quantum machines, which are more accessible in the Noisy Intermediate-Scale Quantum (NISQ) era. This project aims to bridge the gap between these research domains by incorporating quantum-equipped devices, such as quantum edge nodes and quantum clouds, into a Cloud-Edge collaborative computing framework. It highlights the potential of low-qubit quantum machines in resource-constrained environments and introduces them to novel usage scenarios. Moreover, it fosters collaboration between various research and development communities, encompassing cloud-edge computing, cyber-physical systems, and quantum computing. The proposed quantum-classical system emphasizes extensibility, creating a supportive environment for researchers and engineers from these communities, ultimately stimulating innovation and cooperation across these disciplines. It will create opportunities for students to develop their quantum literacy at an underrepresented institution.This project will develop a quantum-equipped Cloud-Edge collaborative computing framework to effectively manage heterogeneous participants, network channels, and quantum noise on resource-constrained devices. The proposed framework consists of three primary components: (i) cloud servers located in remote data centers, providing ample quantum and classical resources; (ii) Edge nodes positioned near end devices with fewer resources than clouds, which can be categorized into two types - quantum-classical and classical-only edges; and (iii) End devices that act as system consumers, possessing minimal resources and potentially equipped with quantum processors. Based on the framework, it provides various modules, including end device registration, resource management, task modeling, and offloading estimation, exploring the potential advantages derived from quantum features such as superposition, entanglements, and teleportation. Specifically, the system builds a client profile for each end device. When a computing job arrives, it models a specific task and generates two execution plans, quantum-classical and classical-only. Based on predefined quantum services that have the potential to provide significant benefits to end devices (e.g., quadratic or exponential speedups), the system predicts the task execution time according to the plans and selects the one that satisfies the client's constraints and maximizes overall system performance. Due to the limited access to quantum machines, the project will implement a distributed quantum-classical Cloud-Edge-CPS simulator to conduct large-scale, cloud-based experiments and support a high degree of heterogeneity. Additionally, it will develop a runtime sampler to study quantum noise effects on resource-constrained NISQ machines, detailing how clouds may help manage inherent noise and errors. Ultimately, this project aims to enhance the performance, efficiency, and capabilities of Cloud-Edge collaborative computing systems by incorporating quantum-equipped devices and investigating their potential benefits in various application scenarios.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.