Cloud Load Balancing Task Scheduling

Project Discription

Increase of digital age raises the calculation cost on servers and cloud. Load balancing resolve this issue by scheduling the jobs and resources. This paper has developed a hybrid model where genetic algorithm BAPSO and Random forest was used for dynamic load balancing. Proposed BAPSO involve Butterfly congnitive and social parameters in each iteration of PSO genetic algorithm. Further use of trained random forest reduce the RPD Total Completion Time for finding of good job sequence. Experiment shows that proposed model has reduced the makespan of the input job matrix by 8.75% as compared to SJFRL. Other parameter were also enhanced by the proposed RFBAPSO model. It was shown that proposed model gives its schedule in less execution time as well. 

Project Sample Image