The proposed method, called LRAPS, tries to estimate short-term CPU utilization of hosts based on their utilization history and is then used to detect overloaded and underloaded hosts as part of live migration process.
Abstract
With tremendous increase in Internet capacity and services, the demand for cloud computing has also grown enormously. This enormous demand for cloud based data storage and processing forces cloud providers to optimize their platforms and facilities. Reducing energy consumption while maintaining service level agreements (SLAs) is one of the most important issues in this optimization effort. Dynamic virtual machine allocation and migration is one of the techniques to achieve this goal. This technique requires constant measurement and prediction of usage of machine resources to trigger migrations at right times. In this paper, we present a dynamic virtual machine allocation and migration method utilizing CPU usage prediction to improve energy efficiency while maintaining agreed quality of service (QoS) levels in cloud datacenters. Our proposed method, called LRAPS, tries to estimate short-term CPU utilization of hosts based on their utilization history. This estimation is then used to detect overloaded and underloaded hosts as part of live migration process. If a host is overloaded, some of the VMs running on that host are migrated to other hosts to avoid SLA violations; if a host is underloaded, all of the VMs in that host are tried to be migrated to other machines so that the host can be powered off. We did extensive simulation experiments using CloudSim to evaluate the efficiency and effectiveness of our proposed method. Our simulation experiments show that our method is feasible to apply and can significantly reduce power consumption and SLA violations in cloud systems.
Cloud computing has seen rapid growth in recent years, leading to a surge in demand for data center services. To meet this demand, data centers deploy a large number of servers, resulting in substantial energy consumption. Virtual Machine Consolidation (VMC) is an effective strategy to reduce energy usage by shutting down underutilized servers while ensuring that Service Level Agreements (SLAs) are maintained. The VMC process comprises four key steps: detecting overloaded hosts, identifying underloaded hosts, selecting virtual machines (VMs), and determining their placement. This research presents the Energy-Efficient Virtual Machine Placement (EEVMP) approach, which aims to optimize resource utilization by selecting suitable destination hosts for migrating VMs based on utilization and resource skewness. The proposed method is evaluated using the CloudSim simulator. Experimental results show that EEVMP consistently outperforms existing placement strategies such as PABFD, IQRMC, PEBFD, ESVMP, and HVMAP in terms of energy efficiency and overall performance.
Dipak Dabhi, A. Kharwar, D. Vadhwani et al.· ITEGAM- Journal of Engineeri...· 0 citations
Cloud computing appears to be an ecosystem with flexible and scalable IT resources. The finding is that cloud services are increasing considerably and resource management is a real problem. This is the case with the static allocation of virtual machines (VMs) that fail to effectively manage multiple workloads in real time. This leads to inefficiencies and high costs. This article proposes a system for dynamic optimization of virtual machines in a cloud to satisfy the multiple and varied requests of users. The main objective is to design a system that dynamically adjusts the number and configurations of VMs according to cloudlet requests, while optimizing performance and costs. This solution allows to learn, scale and remove virtual machines taking into account the variation in demand. It therefore ensures the optimal use of data centre resources.
Aziz Saibou, Onyonkiton Theophile Aballo, Arsene Narcisse Dagba et al.· EPJ Web of Conferences· 0 citations
The findings suggest that HORAM is far better at using resources; fewer tasks are completed, and the total power consumed is lower than with traditional scheduling algorithms, suggesting the suggested architecture is a viable solution to sustainable cloud infrastructure management.
S. Balakrishnan, K. Aravind, ·. T. Veeramani et al.· SN Computer Science· 0 citations
The current invention outlines a Java-driven framework for efficient resource management in cloud data centres using the CloudSim simulation environment. The framework presents a predictive auto-scaling mechanism that examines historical traffic patterns to forecast future workload requirements, allowing for predictive Virtual Machine (VM) al- location rather than traditional fixed threshold-based techniques. Prior to VM migration, the system assesses a Service Level Agreement (SLA) risk factor to avoid performance degradation and potential SLA violations through intelligent power management. A specific Green Scheduler Algorithm dynamically consolidates Virtual Machines by allocating workloads to optimally loaded physical machines based on fore- casted workload conditions. Machines with low utilization are automatically migrated across different power-saving states, such as idle, sleep, and deep sleep modes. This comprehensive framework strikes a balance between energy savings and the preservation of service reliability and Quality of Service (QoS). Simulation results demonstrate the effectiveness of energy savings, improved resource utilization, and SLA compliance, making it suitable for scalable and ecofriendly cloud resource management.
S. Divya, P. Venkadesh, G. Vasunthraa et al.· International Conference on...· 0 citations
The design and development of DynamiCloud is presented, a scalable and computationally efficient multi-objective dynamic resource allocation model for cloud computing that can simultaneously optimize multiple conflicting objectives such as throughput, Service Level Agreement compliance, and power efficiency.
Onwuegbuchulem Gift., Bennett, E.O., Matthias D. et al.· Journal of Artificial Intell...· 0 citations
This paper explores sophisticated Virtual Machine (VM) scheduling approaches in cloud computing and their significance to enhance resource distribution, improve system efficiency, and cost reduction. It provides a recent overviews of key scheduling algorithms, including heuristic, metaheuristic, advanced machine learning-based and hybrid approaches, while assessing their respective strengths, weaknesses and practical applications. The discussion encompasses their applications in managing workloads, optimizing costs, enhancing energy efficiency, improving Quality of Service (QoS) and with a particular focus on scalability and real-time scheduling in cloud settings. Furthermore, the paper analyzes scheduling strategies adopted by major cloud providers through real-world case studies. Ultimately, our analysis identifies critical VM scheduling trade-offs, provides optimization guidelines, and validates the efficacy of hybrid adaptive methods via a new proposed heuristic-machine learning model for dynamic cloud environments.
Chaimae Bahij, Mohamed El Ghmary, Hassan Echoukairi· EPJ Web of Conferences· 0 citations