A New Task Scheduling Algorithm based on Water Wave Optimization for Cloud Computing

Abstract

Nowadays cloud computing provides many benefits for organizations. Businesses can ensure reliable calamity recovery and backup solutions without the spat of tuning them up on a physical machine. For many companies, exploiting complex calamity recovery plans can be an expensive guarantee, and backing up data is time exhaustion. The cloud itself is built in such a way that the data stored more than one time in servers, so that if any server fails, the data is backed up immediately. The capability of accessing data readily is available after handling the failure. However, still, cloud computing resources face many problems such as scheduling problems. This paper tackles the resource scheduling problem and presents a new efficient algorithm, called Improved Water Wave Optimization (IWWO), to address such a problem. The main idea is the enhancement/improvement of the Water Wave Optimization (WWO) algorithm by using reinforcement learning to overcome the local optimality of the conventional WWO during the searching process. The proposed IWWO is implemented in the CloudSim toolkit and evaluated by considering a real data set and a randomly generated data set. The results are compared with the results of the Genetic Algorithm (GA) and Ant Colony Optimization (ACO) algorithm. The obtained results show that the IWWO can solve the resource scheduling with minimum schedule length and a high balance degree.