Article Preview
TopIntroduction
Due to advancements in communication technology and internet usage, as well as their ability to solve complex problems, cloud computing is emerging as a networking technology. Using the internet, cloud users have access to hardware and software resources. An Internet-based computing model called cloud computing allows resources such as software, information, services, storage, and servers to be shared with multiple users (Xu et al., 2018; Kumar et al., 2020). Because of its services to customers, Cloud Computing (CC) is an established business model for distributed computing. The CC model provides for the sharing, allocation, and access of IT resources based on individual needs. CC also offers many services, including Platform-as-a-Service (PaaS), Infrastructure-as-a-Service (IaaS) and Software-as-a-Service (SaaS) (Xue et al., 2018.). They are useful in different domains, including industrial, business, scientific, etc. Amazon, Oracle, HP, IBM, and Apple use cloud computing techniques (Rawat et al., 2020).
A CC platform, in general, suffers from three major problems: Load Balancing, Distributed Framework and Virtualization. In Meeting the demands of cloud users and providers, balancing of load and scheduling of tasks are two major problems in the management of cloud resources (Kumar et al., 2018; Ruan et al., 2019). A load balancer allocates tasks to various machines in such a way that job execution times are reduced and virtual machine performance is monitored (Daming et al., 2020; Polepally et al.,2019). Load balancing's ultimate objective is to lower the highest execution time (maximum Makespan time) whereas raising cloud resource utilization (Kumar & Sharma, 2018). Scheduled tasks can be assigned virtualized resources for a specific period. It canbe done with the help of a cloud service broker and a task scheduling algorithm (Xingjun et al., 2020). Scheduling designates that what tasks will take the lowest time to accomplish. The workload is rising due to the growing set of users in the cloud, but the set of virtual machines (VMs) is remaining the same. Due to the consumption of energy constraints, the set of Virtual machines is reduced by the capacity of PMs. Load balancing and Scheduling tasks ensure that no node is over-or under loaded by the workload (Pradhan et al., 2020; Ragmani et al., 2020).
Static and dynamic load balancing approaches are the two major categories of load balancing approaches (Kruekaew et al., 2020). genetic algorithm (GA), artificial bee colony (ABC), ant colony optimization (ACO) and other optimization algorithms are often used in an existing Complex Load balancing research project. Researchers develop many heuristic and meta-heuristic methodologies such as the jaya algorithm, dragonfly optimization algorithm, and bee colony optimization algorithm and to achieve greater load balancing and task scheduling performance (Priya et al., 2019; Gupta et al., 2021; Milan et al., 2019).