A Systematic Analysis of Load Balance in the Cloud Computing Based on Optimization and Deep Neural Network Algorithms

Advancements in technology have led to the rise of automatic systems that increase the efficiency of business sectors. Cloud computing plays a vital role in business organizations due to its various benefits and services. This survey provides details of existing methodologies used for improving the functioning of task scheduling and task load balancing in the cloud environment. The optimization methods and deep neural network algorithms are considered for the analysis of load balancing. The tasks in the cloud are optimized using optimization algorithms for formulating the most efficient solutions for task scheduling and manage load in the cloud environment. The deep neural networks are trained to learn patterns in task scheduling features using historical data and analyze current task scheduling to maintain load balance in the cloud. The performance of optimization and deep neural network algorithms are evaluated using the performance metrics of execution time, energy consumption, resource utilization, throughput and makespan. The error rates including Mean Square Error (MSE), Root Mean Square Error (RMSE), and Mean Absolute Error (MEA) are also calculated to assess the performance of load balancing.