Cloud is the Latest concept in IT. Users use the resources or services which are provided & managed by the service providers. Users need not to buy the hardware or software which now can be used on rental basis. Workflow represents the cloud application which has different tasks to be executed in an order. Scheduling algorithms are used to assign these tasks to processors and these algorithms decide the cost and time of execution. In this paper, a simple scheduling algorithm has been proposed named Child Based Level-Wise List Scheduling (CBLWLS) algorithm. According to the dependencies CBLWSL calculate priorities of tasks and finds the sequence of task execution and then maps the selected task to the available processors. We perform experiments on Epigenomics workflow structure graphs used in some real applications and their analysis shows that CBLWLS algorithm performed better than the HEFT (Heterogeneous Earliest Finish Time) algorithm, on the parameters of time of execution, execution cost and schedule length ratio.

Cloud computing is a collection of heterogeneous virtualized resources that can be accessed on-demand to service applications. Scheduling large and complex workflows becomes a challenging issue in cloud computing with a requirement that the execution time as well as cost incurred by using a set of heterogeneous cloud resources should be minimizes simultaneously. In this paper, we have extended our previously proposed Bi-Criteria Priority based Particle Swarm Optimization (BPSO) algorithm to schedule workflow tasks over the available cloud resources under given the deadline and budget constraints while considering the confirmed reservation of the resources. The extended heuristic is simulated and comparison is done with state-of-art algorithms. The simulation results show that extended BPSO algorithm also decreases the execution cost of schedule as compared to state-of-art algorithms under the same deadline and budget constraint while considering the exiting load of the resources too.