Growing operational costs and high energy consumption in data centers have increased the need for smart resource allocation and effective energy management. This paper introduces an energy-efficient Virtual Machine (VM) allocation technique that eliminates the need for VM migration while maintaining stable system performance. The proposed approach, named Energy Efficient Virtual Machine Allocation without Migration (EEVMA), aims to lower energy usage, reduce costs, and avoid performance degradation.EEVMA integrates resource optimization techniques, predictive analysis, and real-time workload monitoring to place VMs on servers based on workload patterns, server utilization, and power consumption models. By distributing workloads efficiently and consolidating tasks onto fewer servers, the approach minimizes energy wastage and improves overall efficiency. To validate the effectiveness of EEVMA, we compare its performance with existing VM allocation methods, focusing on metrics such as energy reduction, cost efficiency, and system stability. Experiments using real workload traces and performance indicators show that EEVMA enhances energy savings and service quality without the drawbacks associated with VM migration. The results indicate that data centers can achieve improved sustainability, reduced operational expenses, and better resource utilization through the proposed migration-free allocation strategy. Keywords— Cloud Computing, Energy Efficiency, Resource Management

Energy use in cloud data centers constitutes a large portion of operational cost and carbon footprint. Classic heuristics for autoscaling and resource allocation often fail to adapt to rapidly changing workloads on heterogeneous hardware (CPUs, GPUs, FPGAs). This paper introduces E-RELAY (Energy-aware Reinforcement LEArning for Yield), an RL-based controller that learns to allocate virtualized resources and place workloads across heterogeneous servers to minimize combined energy and SLA-violation costs. E-RELAY models the allocation problem as a Markov decision process where the state captures workload intensity, per-host energy models, temperature, and network contention metrics; the action space includes VM placement, vertical scaling, and migration decisions. To reduce exploration risk, we incorporate a model-based planning layer that simulates candidate actions with a learned surrogate energy model. We evaluate E-RELAY on realistic, public workload traces and a production-scale simulator (including mixed CPU/GPU tasks). Results show 12–25% energy savings vs. rule-based baselines for equal SLA attainment, and 20–35% fewer SLA violations under sudden load spikes. We also analyze performance under heterogeneity, showing larger relative gains when accelerators are present. Finally, we discuss deployment considerations: cold-start bootstrapping, safety constraints for migration decisions, and how learned policies generalize across data center topologies. E-RELAY demonstrates that combining RL with accurate energy surrogates and conservative planning yields practical energy savings while respecting production constraints.

Cloud computing has become the backbone of digital ecosystems, but growing workloads intensify challenges in resource optimization, virtual machine (VM) migration, and security assurance. Existing studies often address these issues in isolation, limiting their practical applicability. This paper presents a unified framework that integrates three complementary components: (i) an Improved Modified Particle Swarm Optimization (IMPSO) algorithm with adaptive inertia scheduling and dynamic mutation control, which outperforms IPSO in convergence speed and load distribution accuracy; (ii) a machine learning–assisted hybrid live VM migration method with dirty-page clustering and workload prediction to minimize downtime; and (iii) a blockchain-enabled secure migration layer to ensure tamper-proof and auditable state transfer. The revised version of this study includes statistical validation (confidence intervals, t-tests) and attack simulation experiments (e.g., man-in-the-middle and replay attacks) to ensure methodological rigor and realistic security assessment. Experimental results on a real XenServer testbed show that the proposed system improves response time by ~30%, reduces migration downtime by ~60%, and ensures 100% migration integrity with ?15% security overhead. Overall, this work represents among the first unified frameworks that jointly optimize resource allocation, downtime reduction, and blockchain-based security in a practically validated, end-to-end cloud migration environment.

An integrated technique for securing large virtual machine migration

GrC-VMM: An intelligent framework for virtual machine migration optimization using granular computing

A Bayesian neural network study for virtual machine migration within cloud environment

The surging energy requirements and the impact on the environment of the cloud data center suggests the necessity of establishing intelligent resource management plans to facilitate sustainability. The proposed hybrid framework which incorporates hybrid dynamic virtual machine (VM) consolidation based on the idea of artificial intelligence (AI) with the concept of renewable-aware differences in this paper to decrease energy and carbon emission in cloud computing environment. The consolidation module employs reinforcement learning to dynamically locate and move VMs based on real time workload and server utilization patterns. Concurrently, a renewable-sensitive scheduler predicts availability of solar energy and wind energy and executes tasks around green energy highlights. The framework is tested with the help of CloudSim Plus, actual Google Cluster traces, and synthetic renewable traces. The findings indicated that there was a decrease of energy of 37 percent and carbon emission decrease of 46.7 percent over the traditional models with a high SLA compliance along with minimizing the VM migration overhead. The given solution has a great potential to become an energy-efficient and environmentally-friendly cloud infrastructure that does not worsen performances.

In the contemporary data centers, virtual machine migration is a crucial technique for gaining scalability through optimum resource use. The majority of earlier publications focused on how to move virtual machines in Cloud environments with high traffic needs. It mainly concentrated on the internet activity between virtual computers and the energy required by those machines. None of the preceding algorithms gave consideration to the client's experience, such as the wait time they experienced to have their services handled. Among the most significant problems that Cloud technology is now experiencing is load balancing. All the nodes should receive an equitable distribution of the load. In dynamic and diverse contexts, dynamic algorithms produce superior outcomes. This research suggests an enhanced particle swarm optimization approach for Cloud Computing job scheduling optimization. First, a scheduling model utilizing an upgraded particle swarm technique is suggested based on the Cloud Computing scheduling algorithm concept to prevent the optimization approach from devolving into local optimization. Three load balancing algorithms are used for comparison: the Honeybee Foraging Behavior Load Balancing Algorithm, the Throttled Load Balancing Algorithm, and the ESCE (Equally Spread Current Execution) Task Allocation Algorithm. It is demonstrated through the simulation results that the suggested TSPSOA is superior to the other state-of-the-art algorithms.

The article discusses various aspects of virtual machine migration. The process of moving them between different hosts, data storages, or even between cloud environments. It can be performed in two modes: cold migration, when the virtual machine is previously turned off, and live (dynamic) migration, which occurs without stopping its operation. Thanks to live migration, it is possible to transfer active virtual machines, for example, between servers within a cluster, without interrupting the provision of services. To organize such a process, special administration tools are used, in particular Hyper-V Manager or System Center Virtual Machine Manager from Microsoft Learn.

: This paper discusses virtual machines (VM) and their use in server technology. This study focuses on the use of machine learning (ML) to schedule live migrations of VMs. The authors conducted a systematic literature review (SLR) to gather evidence of ML research in server migration. The study found that there is a lack of research in this area, and quantitative research can be conducted to explore the potential of ML in terms of server migration. The paper also presents a selection of paper criteria defined for the study, including the exclusion and inclusion criteria, quality assessment, and quantity assessment. The authors retrieved reliable and related papers using the definition of the paper selection criteria and keywords. The SLR method is not discussed in all papers, and the authors want to develop the title into SLR format to produce high-quality papers on live migration VM machine learning. The paper also includes a journal review that discusses the theory of graph team infra and the scheduling algorithm. The authors also present their research questions, which include the definition of virtual machine, live migration, and its application. The paper includes a list of references that discuss various aspects of VMs, including migration strategies, scheduling methods, and self- management of virtual network resources. The authors conclude that there is a need for further research in the area of ML in server migration, and quantitative research can be conducted to explore the potential of ML in this field.

Reducing energy consumption and optimizing resource usage are essential goals for researchers and cloud providers managing large cloud data centers. Recent advancements have demonstrated the effectiveness of virtual machine consolidation and live migrations as viable solutions. However, many existing strategies are based on immediate workload fluctuations to detect host overload or underload and trigger migration processes. This approach can lead to frequent and unnecessary VM migrations, resulting in energy inefficiency, performance degradation, and service-level agreement (SLA) breaches. Moreover, traditional time series and machine learning models often struggle to accurately predict the dynamic nature of cloud workloads. This paper presents a consolidation strategy based on predicting resource utilization to identify overloaded hosts using novel hybrid value trend generative adversarial network (VTGAN) models. These models not only predict future workloads but also forecast workload trends (i.e., the upward or downward direction of the workload). Trend classification can simplify the decision-making process in resource management approaches. We perform simulations using real PlanetLab workloads on Cloudsim to assess the effectiveness of the proposed VTGAN approaches, based on value and trend, compared to the baseline algorithms. The experimental findings demonstrate that the VTGAN (Up current and predicted trends) approach significantly reduces SLA violations and the number of VM migrations by 79% and 56%, respectively, compared to THR-MMT-PBFD. Additionally, incorporating VTGAN into the VM placement algorithm to disregard hosts predicted to become overloaded further improves performance. After excluding these predicted overloaded servers from the placement process, SLA violations and the number of VM migrations are reduced by 84% and 76%, respectively, compared to THR-MMT-PBFD.

With the advancement of rapid communication and information technology, the demand for efficient and reliable cloud computing services has gradually grown. Power consumption at the cloud data centers becomes a challenging issue for such demanding requirements since user workloads are mostly unpredictable and exhibit a dynamic nature. This paper proposes a hybrid technique for energy optimization in cloud data centers. Neuro-fuzzy networks, integrated into the workload prediction mechanism, are used together with the Ant Colony Optimization (ACO) framework for virtual machine (VM) migration and placement. The main goal of this work is the minimization of the number of active servers by fulfilling all user requests without increasing energy consumption. The proposed model leverages neuro-fuzzy networks for accurate workload predictions to achieve efficient real-time resource allocation and VM migration strategies that minimize energy waste. The empirical results indicate that the proposed approach minimizes energy consumption with a low request rejection rate. Key contributions of this work are as follows: the efficiency of resource allocation has been enhanced; the real-time load is better predictable; operational cost is reduced; and, consequently, the profitability of cloud service providers has increased. This framework further proposes to help increase customer satisfaction and competitiveness in the cloud market by ensuring that cloud services are delivered reliably and efficiently. The work is applied to the very important sustainability issue of cloud computing, providing a robust framework suitable for the dynamic and nonlinear behaviors of cloud environments. The results underlined the potential of predictive models combined with optimization algorithms for significant energy savings and operational efficiency in cloud data centers.

Virtual machine (VM) placement in cloud datacenters is a complex multi-objective challenge involving trade-offs among energy efficiency, carbon emissions, and network performance. This paper proposes NCRA-DP-ACO (Network-, Cost-, and Renewable-Aware Ant Colony Optimization with Dynamic Power Usage Effectiveness (PUE)), a bio-inspired metaheuristic that optimizes VM placement across geographically distributed datacenters. The approach integrates real-time solar energy availability, dynamic PUE modeling, and multi-criteria decision-making to enable environmentally and cost-efficient resource allocation. The experimental results show that NCRA-DP-ACO reduces power consumption by 13.7%, carbon emissions by 6.9%, and live VM migrations by 48.2% compared to state-of-the-art methods while maintaining Service Level Agreement (SLA) compliance. These results indicate the algorithm’s potential to support more environmentally and cost-efficient cloud management across dynamic infrastructure scenarios.

The use of cloud computing, particularly virtualized infrastructure, offers scalable resources, reduced hardware needs, and energy savings. In Ethiopian public hospitals, the lack of integrated healthcare systems and a national data repository, combined with existing systems deficiencies and inefficient traditional data centers, contribute to energy inefficiency, carbon emissions, and performance issues. Thus, evaluating the energy efficiency and performance of a cloud-based model with various workloads and algorithms is essential for its successful implementation in healthcare systems and digital health solutions. The study experimentally evaluates a cloud-based model's energy efficiency and performance for smart healthcare systems, employing descriptive and experimental designs to simulate cloud infrastructure. Simulations are conducted on diverse workloads in CloudSim using power-aware (PA) algorithms (along with VmAllocationPolicy and VmSelectionPolicy), and dynamic voltage frequency scaling (DVFS). Results reveal that the number of VMs and their migrations significantly impact energy consumption, with some algorithms achieving notable energy savings. Lr/Lrr-based algorithms are particularly energy-efficient, with LrMc and LrrMc saving 29.36% more energy than IqrMu at 55 VMs, and LrrRs saving 30.20% more at 1,765 VMs. DVFS adjusts energy consumption based on the number of VMs, while non-power-aware (NPA) consumes maximum energy based on hosts, regardless of the number of VMs. VM migrations, energy consumption, and average SLAV are positively correlated, while SLA is negatively correlated with these factors. In PlanetLab, energy consumption and average SLAV show a strong positive correlation (0.956) at Workload6, while SLA at Workload2 and average SLAV at Workload1 show a weak negative correlation (-0.055). Excessive migrations can disrupt the system's stability/performance and cause SLA violations. Task completion time is influenced by VM processing power and cloudlet length, being inversely proportional to VM processing power and directly proportional to cloudlet length. Overall, the findings suggest that cloud virtualization and energy-efficient algorithms can enhance healthcare systems performance, patient care, and operational sustainability.

With the evolution of cloud computing (CC) technologies, there is a growing insistence for the maximum utilization of cloud resources, therefore increasing the computing power consumption of cloud’s systems. Cloud’s Virtual Machines (VMs) consolidation imparts a practical mechanism to minimize energy consumption of cloud Data Centers (DC). Efficient consolidation and migration of VM in the absence of infringing Service Level Agreement (SLA) can be arrived at by making decisions proactively based on cloud’s future workload prediction. Efficient load balancing, another major issue of CC also depends on accurate forecasting of resource usage. Cloud workload traces reveal both periodic and non-periodic patterns with the unexpected peak of load. As a result, it is very demanding for the prediction models to accurately anticipate future workload. This prompted us to propose a method called, Attention Bidirectional Gated and Weight-adaptive Sparrow Search Optimization (ABiG-WSSO) to accurately forecast future workload with minimal makespan and overhead. The proposed ABiG-WSSO method includes Attention Bidirectional Gated Recurrent Unit (ABiGRU) and Weight-adaptive Sparrow Search Optimization (WSSO). Attention Bidirectional Gated Recurrent Unit (ABiGRU) is initially designed that along with the use of Bidirectional Gated Recurrent Unit (BiGRU) and adaptation of attention mechanism aids in predicting future cloud load requirements accurately. Next, Weight-adaptive Sparrow Search Optimization (WSSO) algorithm is employed in fine-tuning the parameters of the ABiGRU model for accurate and optimal load balancing performance. The WSSO algorithm is applied to optimize ABiGRU model hyperparameters (i.e. learning rate), to enhance its prediction accuracy. Comprehensive simulations are carried out using the gwa-bitbrains dataset to verify the efficiency of the proposed ABiG-WSSO method in boosting the distribution of resources and cloud load balancing. The proposed method achieves comparatively better results in terms of better makespan time, energy consumption, associated overhead and throughput.

Financial cloud computing is based on the virtualization technology paradigm, which has recently gained prominence in the information technology (IT) industry. The resource allocation in the existing system is not guaranteed, and convergence problems occasionally cause the process to move slowly. Consequently, there has been a notable decline in the overall efficacy of financial cloud computing. In this work, new methods are presented to improve the financial cloud. The three main stages of the proposed system are resource allocation, load balancing and cost-effective virtual machine (VM) migration. To enhance load balancing, this study employs the Enhanced Weighted Round Robin Algorithm (EWRR) technique. Load balancing is achieved by shifting workloads from overloaded to underloaded nodes. The Differential Evaluation-based Bat Algorithm (DEBA) efficiently chooses more optimal resources to accomplish the optimal resource allocation in the financial cloud. Resolute Support Vector Machine (RSVM) technique is utilized to Provide Cost-effective VM migration. The simulation's findings show that the recommended DEBA-EWRR algorithm performs better than existing techniques due to advancements in throughput, Makes pan, Fitness score and resource utilization.

ABSTRACTThe increase in popularity and demand for cloud services has caused a huge growth of cloud data centers, and this has caused the challenge of energy management in data centers. Virtual Machine (VM) consolidation is a critical process aimed at optimizing resource utilization and minimizing energy usage. VM consolidation with the turnoff of underloaded hosts and reducing the load of overloaded hosts establishes a balance between energy consumption and SLA violations. In fact, the consolidation process includes three sub‐problems: determining overloaded and underloaded hosts, VM selection in overloaded hosts, and finding a new destination for VMs that will be migrated (VM placement). This paper introduces an entropy‐based approach to VM selection and placement to improve efficiency in cloud data centers. Entropy is a quantifiable characteristic often linked to disorder, randomness, or unpredictability. By leveraging entropy as a measure of workload distribution and uncertainty, the proposed method effectively predicts future resource demands, enabling informed decisions that enhance energy efficiency and reduce SLA violations. A key advantage of this approach is the significant reduction in the number of VM migrations, which decreases overhead and minimizes potential service disruptions. Experimental results demonstrate that our entropy‐based method outperforms the VM consolidation process in terms of energy consumption, SLA compliance, and system stability. The findings suggest that this approach offers a more sustainable and cost‐effective solution for managing cloud resources, contributing to the development of efficient and reliable cloud computing environments.

Task scheduling with enhanced VM migration using SM-PCCTSA with SKD-RBM

ABSTRACTCloud computing is an innovative technology that provides computing services over the internet and replaces the requirement to own physical hardware or software. Security threats present a wide range of risks to cloud computing, and a security threat defense plays a significant role in cloud computing. Virtual machines (VM) serve as the backbone, providing flexible and scalable resources for running and storing data. Moving Target Defense (MTD) and Blockchain enhance security and privacy by reducing the chances of successful attacks and minimizing the impact of security attacks. To address these issues, we propose integrating MTD and blockchain technologies within the cloud computing environment named Hybrid Secure Onlooker (HSO). The proposed work involves several entities, including Cloud Users (CUs), Centralized Subnet Manager (CSM), Distributed Group Manager (DGM), Consortium Block Module (CBM) and Private Block Module (PBM). Initially, we perform Multi‐Factor Authentication (MFA) to establish secure communication and to avoid malicious traffic. Followed by this, we utilize the Komoda Miliphir optimization (KMO) algorithm to perform CUs' task scheduling based upon the task types, task sensitivity, and task size. Entrenched in the scheduled tasks, the CSM performs classification and grouping of cloud VMs, assigning them to their capacity, security protocols, and availability, utilizing the Residual Flowed Capsule Network (RFC‐Net). The grouped subsets are overseen and managed by the DGM, which handles MTD operations such as virtual switch placement and VM migration within the subsets. Finally, the transactions are stored in the hybrid blockchain layer with CBM and PBM to ensure privacy and security. The is the implementation tool for realizing the proposed HSO model. The proposed model can be examined based on several metrics with state‐of‐the‐art work comparisons. The results show that the proposed HSO model outperforms the state‐of‐the‐art models.

Optimizing cloud resource management with an IoT-enabled optimized virtual machine migration scheme for improved efficiency