Search

The associated cloud task is a typical application mode of scientific workflow, in which the scheduling is different from the traditional independent tasks. For solving the scheduling optimization problem of this cloud workflow task, a workflow scheduling algorithm meeting budget constraint is proposed. The proposed algorithm divides the best mapping scheme between tasks and instances into four stages: workflow structure division, budget distribution, task selection and instance selection. The workflow structure division is to divide all tasks into different levels according to the feature of workflow structure. The budget distribution is to allocate the user-defined budget to each defined level. The task selection is to give priority to tasks based on the sum of upward rank and downward rank , and form the list of ready tasks, The instance selection is to select the best execution instance to schedule tasks meeting the available sub-budget. We test the performance of the algorithm under different types of workflow structures and constraint strict degree by the simulation experiments. The results show that the execution efficiency and the scheduling success rate of our algorithm are better than the same type of other algorithms, it has better feasibility. [ABSTRACT FROM AUTHOR]