Abstract: This study investigates a quantum-inspired optimisation framework for energy-aware virtual machine (VM) placement in cloud data centres based on a Quadratic Unconstrained Binary Optimisation (QUBO) formulation. The proposed approach encodes the VM–host assignment problem into a QUBO Hamiltonian that jointly minimises energy consumption and penalises service level agreement (SLA) violations and capacity constraint violations. Quantum-inspired solvers implemented using TensorFlow Quantum and Qiskit are employed to approximate low-energy configurations, which translate to near-optimal placement decisions. The framework is evaluated on realistic workload traces calibrated to empirical characteristics of production cloud environments and executed on a heterogeneous testbed of physical servers. Experimental results demonstrate that the quantum-inspired QUBO optimisation achieves up to 17.8% lower energy consumption compared with a power-aware best-fit decreasing heuristic, while simultaneously reducing SLA violations by 6.6%. The results further show that the QUBO-based approach provides more stable server utilisation profiles and a higher degree of VM consolidation without incurring excessive migration overhead. These findings indicate that quantum-inspired optimisation, even when executed on classical hardware through TensorFlow Quantum and Qiskit simulators, offers a promising direction for improving the sustainability and efficiency of large-scale cloud data centres. Keywords Quantum-inspired optimisation; QUBO; energy-aware VM placement; TensorFlow Quantum; Qiskit; cloud data centres.
Abstract: This study investigates a quantum-inspired optimisation framework for energy-aware virtual machine (VM) placement in cloud data centres based on a Quadratic Unconstrained Binary Optimisation (QUBO) formulation. The proposed approach encodes the VM–host assignment problem into a QUBO Hamiltonian that jointly minimises energy consumption and penalises service level agreement (SLA) violations and capacity constraint violations. Quantum-inspired solvers implemented using TensorFlow Quantum and Qiskit are employed to approximate low-energy configurations, which translate to near-optimal placement decisions. The framework is evaluated on realistic workload traces calibrated to empirical characteristics of production cloud environments and executed on a heterogeneous testbed of physical servers. Experimental results demonstrate that the quantum-inspired QUBO optimisation achieves up to 17.8% lower energy consumption compared with a power-aware best-fit decreasing heuristic, while simultaneously reducing SLA violations by 6.6%. The results further show that the QUBO-based approach provides more stable server utilisation profiles and a higher degree of VM consolidation without incurring excessive migration overhead. These findings indicate that quantum-inspired optimisation, even when executed on classical hardware through TensorFlow Quantum and Qiskit simulators, offers a promising direction for improving the sustainability and efficiency of large-scale cloud data centres. Keywords Quantum-inspired optimisation; QUBO; energy-aware VM placement; TensorFlow Quantum; Qiskit; cloud data centres.
The digital transformation that Telecommunications and ICT domains are crossing today, is posing several new challenges to Telecom Operators. These challenges require solving complex problems such as: dimensioning and scheduling of virtual/real resources in data centers; automating real-time management/control and orchestration of networks processes; optimizing energy consumption; and overall, ensuring networks and services stability. These problems are usually tackled with methods and algorithms that find suboptimal solutions, for computational efficiency reasons. In this work, we consider a Virtual Data Center scenario where virtual machine consolidation must be performed with joint minimization of network/servers power consumption. For this scenario, we provide an ILP model, the equivalent binary model and the steps towards the equivalent Quadratic Unconstrained Binary Optimization (QUBO) model that is suitable for being solved by means of quantum optimization algorithms. Finally, we compare the computational complexity of classical and quantum solvers from a theoretical perspective.
Cloud computing has transformed modern digital infrastructure by enabling on-demand access to scalable computational resources. However, the increasing complexity of dynamic workloads and the need for efficient resource allocation present persistent challenges in maintaining performance, cost efficiency, and energy sustainability. This paper introduces a Quantum-Driven Optimization (QDO) framework, a hybrid quantum-classical approach designed to enhance cloud resource allocation by integrating quantum computing techniques with classical optimization methods. The proposed framework leverages Quantum Annealing (QA) and the Variational Quantum Eigensolver (VQE) to optimize resource distribution, minimizing energy consumption and operational costs while maximizing throughput and utilization efficiency. Experimental evaluations demonstrate that the QDO framework achieves a 27% improvement in resource utilization, a 34% reduction in operational costs, and a 21% enhancement in task completion time compared to traditional heuristic-based approaches. Additionally, the hybrid model reduces Service Level Agreement (SLA) violations by 18%, ensuring robust Quality of Service (QoS) for cloud users. The framework employs classical algorithms for preprocessing and decision-making while delegating complex optimization tasks to quantum solvers, ensuring scalability across diverse cloud environments. This study highlights the transformative potential of hybrid quantum-classical computing in addressing cloud resource allocation challenges. The results indicate significant improvements in energy efficiency, cost-effectiveness, and system responsiveness, making the QDO framework a viable solution for next-generation cloud infrastructures. Future research directions include extending the framework to multi-cloud architectures and investigating advanced quantum algorithms for further optimization gains.
Quantum Optimization (QO) is emerging as a promising approach to tackle hard combinatorial optimization problems by leveraging the unique computational properties of quantum hardware. In recent years, QO has gained traction in Machine Learning (ML), where problems such as clustering and vector quantization can be naturally expressed as Quadratic Unconstrained Binary Optimization (QUBO) problems. These ML applications often involve complex, high-dimensional data, which directly influences both the structure and scale of the resulting QUBO formulations. However, current quantum devices fall under the category of Noisy Intermediate-Scale Quantum (NISQ) hardware, which imposes significant limitations on qubit counts and small error rates. This dissertation investigates the impact of data complexity and data scale—as encountered in ML settings—on the effectiveness and feasibility of QO on NISQ devices. We propose novel, hardware-aware methods to overcome these challenges. The first part of this thesis focuses on the effect of data complexity for QO. We analyze how problem structure—particularly when QUBO formulations are derived from ML tasks—affects quantum solvability. We identify the spectral gap as a key factor in the success of adiabatic quantum algorithms and provide a theoretical and empirical analysis linking data-induced complexity to optimization performance.To address noise sensitivity and limited precision, we introduce a preprocessing framework based on a principled Branch-and-Bound algorithm that reduces dynamic range while preserving global optima. This improves robustness on both quantum annealers and classical solvers. In the second part, the thesis addresses challenges introduced by data scale. First, we propose a recursive Divide-and-Conquer strategy that breaks down large QUBO problems into tractable subproblems, demonstrated through an application to Bundle Adjustment in 3D scene reconstruction. Second, we introduce a variable and constraint generation method inspired by column generation for Integer Linear Programming, enabling dynamic QUBO size growth. This technique is successfully applied to Multi-Agent Pathfinding, where it scales well on NISQ devices while maintaining compatibility with classical solvers. Finally, we explore QUBO reformulations that reduce problem size through compact encodings and cyclic expansion, enabling the use of quantum hardware for tasks such as FPGA placement in chip design. Across all contributions, the developed algorithms are evaluated both theoretically and empirically, with experiments conducted on real quantum and quantum-inspired hardware. These methods significantly improve solution quality, scalability, and hardware compatibility, paving the way for more effective use of QO in practical settings. While tailored for NISQ era constraints, the results remain relevant for fault-tolerant quantum computing, offering insights into the long-term viability of QO across hardware generations.
As systems become more complex and large-scale, there is a need for more effective scheduling algorithms to maximize resource usage and minimize job completion times. The older methods of scheduling like First-Come-First-Serve (FCFS) and Round Robin (RR), although easier to compute, tend to perform poorly in dynamic scenarios because they cannot manage intricate constraints and multi-objective optimization efficiently. As traditional methods approach their limit of performance, quantum computing presents a potentially fruitful avenue for conquering combinatorial challenges based on its ability to examine many states at once. This work proposes a quantum-aided scheduling framework that employs Quadratic Unconstrained Binary Optimization (QUBO) formulations and quantum solvers for the purpose of improving job scheduling in distributed environments. The original contribution is the marrying of quantum optimization (with IBM Qiskit's QAOA and D-Wave's hybrid solvers) with traditional distributed computing frameworks such as Apache Spark. Google Cluster Workload Traces are employed to mimic real-world situations, and jobs are described by CPU, memory requirements, and arrival times. Once the scheduling problem has been formulated into a QUBO model, quantum algorithms produce the optimized job-to-machine assignments. The outcomes are compared against baseline classical schedulers on performance metrics including make span, throughput, and resource utilization. The experiment demonstrates that quantum-assisted scheduling significantly improves makespan and enhances resource utilization, particularly under high workloads. It also provides evidence of the feasibility of integrating quantum solvers with existing big data infrastructures.
Cloud computing has fundamentally transformed digital infrastructure by providing on-demand access to scalable computational resources. Yet the inherent unpredictability of workloads alongside surging demands for efficient resource distribution continues to present substantial operational hurdles. This investigation introduces a Quantum-Driven Optimization framework that synergistically combines quantum and classical computational paradigms to improve resource allocation efficacy within cloud ecosystems. The proposed methodology employs Quantum Annealing alongside Variational Quantum Eigensolver algorithms to optimize resource distribution through simultaneous minimization of energy expenditure and operational overhead while maximizing throughput and utilization rates. Empirical validation demonstrates that our framework achieves measurable improvements over conventional heuristic methodologies, specifically showing enhanced resource utilization, reduced operational costs, and accelerated task completion times. The hybrid architecture effectively distributes computational burden by delegating preprocessing and decision-making functions to classical systems while quantum processors tackle inherently complex optimization problems. The solution demonstrates scalability and adaptability across diverse cloud deployment scenarios while exhibiting reduced Service Level Agreement violations, thereby ensuring superior Quality of Service for end-users. This research contributes meaningful insights regarding the practical implementation of hybrid quantum-classical approaches for addressing contemporary resource allocation challenges within cloud computing infrastructures.
This paper proposes an integrated QUBO-based quantum-annealing optimization framework for next-generation cloud computing. First, for resource-demand forecasting, we blend Autoregressive (AR) models with Quadratic Unconstrained Binary Optimization (QUBO) and solve the resulting problem via Simulated Annealing; monthly-scale tests reveal superior trend-capture accuracy compared with classical predictors. Second, we embed quantum optimization into Support Vector Machines by casting kernel selection and penalty-parameter tuning as QUBO instances; evaluations on the Iris benchmark—and on synthetically enlarged variants—show marked gains in both classification accuracy and training speed, especially under large-data regimes. Third, we extend the same paradigm to deep learning: a QUBO-formulated quantum-annealing routine optimizes weight initialization and layer-wise learning rates in convolutional networks for image classification, yielding higher top-1 accuracy and reduced wall-clock training time. Across all tasks the hybrid quantum-classical solver consistently outperforms its classical counterparts while maintaining linear scalability in cloud environments. These results show the practical potential of quantum - enhanced algorithms for resource prediction, classification, and deep learning, and pave the way for wider use of quantum technologies in cloud - based AI services.