A Probabilistic Genetic Algorithm Approach to Efficient Task Scheduling in Cloud Environments

Task scheduling is a significant element for any type of distribution system since it routes tasks for precise resources for execution in cloud computing systems. Furthermore, search-based scheduling algorithms select tasks via various methods, resulting in extensive execution times on mixed distributed systems. Consequently, task prioritizing results in a blockage in these systems. Therefore, selecting tasks with the minimum execution time via robust algorithms is applicable. Hence, the genetic algorithm (GA) is among the evolutionary methods utilized to resolve difficulties rapidly. The traditional algorithms include the estimation-of-distribution algorithm (EDA), GA, and EDA-GA algorithms. These algorithms are used to optimize the resource utilization with less time complexity. However, the load balancing process that distributes the tasks over multiple computing resources and task scheduling within cloud computing increases the efficiency by optimizing the resource utilization, which ensures high scalability and availability. Hence, the probability of discovering a reorganizing-based GA has the intention to optimize the task allocated in the virtual machines (VMs), which reduces the completion time and execution cost when optimizing the resource utilization. By the impact of genetic operators such as selection, crossover and mutation algorithms effectively direct the solution space to determine optimal tasks. Furthermore, the performance of the algorithm is calculated via the Simpy toolkit, which indicates the capacity to outperform existing scheduling techniques. Moreover, the experimental results reveal that at 1000 iterations, the proposed model has a completion time of 10,841 ms compared with the other algorithms. Hence, it has a significant effect on efficacy and balances the competing time. This study provides functional effects to improve task scheduling approaches in cloud computing, providing a valuable structure for future research to optimize cloud resource management.