Workflow Task Scheduling Hybrid swarm intelligence in cloud Computing

Abstract Users may access virtual, scalable, and dynamic resources using cloud computing, which is a novel technology that charges them only for the resources they use. This technology is reliant on a network to function. Several services, like as e-commerce apps, use the whole network as a result of the sheer scope of the network. Cloud computing has progressed through three phases in terms of practical development: distributed computing, parallel computing, and grid computing. Cloud computing is a cutting-edge technology that instals networks of computers, many of which are situated in faraway locations, to conduct operations on massive amounts of data in real time. Cloud computing makes use of a workflow model to describe various scientific and web-based applications in the cloud. On the heterogeneous cloud environment, one of the most difficult problems to solve is the scheduling of massive workflows of jobs that adhere to scientific criteria. Other challenges are unique to public cloud computing, such as security and privacy. For example, customers must be happy with quality of service (QoS) metrics such as scalability and dependability, as well as the requirement to maximise end-user resource consumption rates, among other things. This study examines scheduling algorithms that are based on the Water cycle optimization principle. Specifically, it is intended to aid users in determining the most appropriate quality of service consideration for big workflows in infrastructure as a service Shabina Ghafir et al. (IaaS) cloud applications and mapping tasks to available resources. In addition, the scheduling methods are classified based on the variation of the WCA algorithm that is used to implement them. For the purpose of comparing and identifying the potency of the suggested system, the current round-robin (RR), ALO, and PSO approaches have been selected. The results showed that the suggested approach reduces the cost by 9.8 percent for GA-PSO, 10 percent for PSO, 20 percent for ALO, 30 percent for RR, and 12 percent for GA by using a combination of these techniques. The suggested solution decreases load balancing and makespan by 8 percent compared to GA-PSO, 10 percent compared to ALO, 20 percent compared to PSO, 35 percent compared to RR, and 45 percent compared to GA. Furthermore, the performance of energy consumption and reliability are both satisfactory.