Dynamic Task Scheduling using Trained Neural Network and Genetic Algorithm

Abstract

This paper focus on allocation of cloud resources where two models were developed for this work. First was TLBO (Teacher Learning Based Optimization) genetic algorithms which find the correct position for the process to execute. Here some information used for analysis are total number of machines, memory, execution time, etc. So, outcome of the selected training process sequence were used as training input to the Convolutional Neural Network for learning. Here training was done in such a way that all set of features were utilized in pair with their process requirement and current position. For increasing the reliability of the work whole experiment was done on real dataset. Result shows that proposed GNNLB model has overcome various evaluation parameters on different scale as compared to previous approaches adopt by researchers.