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

点击翻译按钮获取中文摘要

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.