An Energy Efficient Virtual Machines Placement in Cloud Datacenters using Adaptive Greedy Dingo Optimization Algorithm

Abstract

More dependable service storage and a lower cost model for different forms of data storage are the primary objectives of the cloud data storage process, without taking infinite scalability into account. There are requirements that cloud users must fulfill because of stored procedures. However, since it impacts the data's quality and integrity, keeping a lot of data is crucial. Utilizing a hybrid heuristic technique, we suggest a cloud storage model that is dependable and effective to surpass these difficulties. Managing deployment and typical limits in cloud environments, as well as optimizing data storage, are the primary objectives of this system. Priorities are set for generic device capacity restrictions and data allocation rules. To overcome these constraints, we optimize the cloud data storage components using the Adaptive Greedy Dingo Optimization Algorithm (AGDOA). Ineffective virtual machine deployment (VMP) and the sharing of shared physical systems by several users are the outcomes of this, which raises interworking costs, wastes resources, consumes excessive power and creates security risks. A novel system called Secure and Multipurpose Virtual Machine Deployment (SM-VMP) is provided with efficient virtual machine migration capabilities to tackle the aforementioned difficulties. In addition to minimizing reciprocal communication delays and ensuring energy-efficient resource deployment between virtual machines, the suggested system places a strong emphasis on the timely and secure execution of user applications. Whale evolutionary optimization and non-dominated sorting-based genetic algorithms serve as inspiration for the proposed ISOA (Improved Seagull Optimization Algorithm), which is used to achieve VMP. The results are examined and contrasted with those obtained from previous approaches after the performance has been confirmed. Thus, for accurately storing cloud data, the suggested paradigm yields optimal outcomes.