Flexible Genetic Algorithm Operators for Task Scheduling in Cloud Datacenters

Cloud computing is the backbone of the modern information technology industry. Due to the increase in internet usage, social media, and smart phones, a large amount of data is producing. Cloud datacenters can provide resources to handle this data. If data is not properly handled, it can cause overhead on cloud servers and can increase operational costs. Genetic algorithm is used to solve scheduling problem efficiently, but they take a lot of time to find an optimal solution. In this paper, we proposed Flexible Genetic Algorithm Operators (FGAO) for Task Scheduling in Cloud Datacenters. This algorithm changes crossover and mutation operators according to the quality of scheduling solutions. Instead of giving a fixed stopping criteria algorithm uses flexible crossover and mutation operators as a stopping criterion. Experimental results show that the proposed FGAO algorithm reduces 40% execution time and 33% iterations as compared to the genetic algorithm.