Efficient Optimization in Dispensation of Cloud applications using Novel Stochastic Gradient with Forward Back Propagation Algorithm

The effective use of cloud server is a major challenge in the recent days. To handle the computation speed, space complexity, network accessibility is playing vital role in achieving the efficiency of cloud server. The objective is to obtain the energy efficiency in cloud servers using a novel stochastic gradient with forward back propagation algorithm. Two sample groups with number of iterations 20 each have been tested with G-power of 80% with the total sample size of 360, divided into two groups with Group 1=180, Group 2=180 and an independent sample t-test has been done. To improve the accuracy of energy optimization, a novel stochastic gradient with forward back propagation algorithm is proposed and compared with the machine learning algorithm. The results prove that the novel stochastic gradient with forward back propagation algorithm has a high accuracy of 92.1%, which is significantly better than the other machine learning algorithms. The level of significance p=0.02 (p<0.05) shows that there is a significant difference between these two algorithms. Thus, the cloud server optimization is achieved using stochastic gradient with forward back propagation algorithm through various testbeds.