Task Duplication-Based Scheduling Algorithm for Budget-Constrained Workflows in Cloud Computing

Workflow scheduling is crucial to the efficient operation of cloud platforms, and has attracted a lot attention. Up now, many algorithms have been reported schedule workflows with budget constraints, so as optimize workflows' makespan on resources. Nevertheless, hourly-based billing model in computing an ongoing challenge for workflow that easily results higher or even infeasible solutions. Besides, due data constraints among tasks, there must be idle slots Few works adequately exploit these duplicate tasks' predecessors shorten their completion time, thereby minimizing workflow's while ensuring its constraint. Motivated by these, we propose task duplication based algorithm, namely TDSA, budget-constrained platforms. In two novel mechanisms are devised: 1) dynamic sub-budget allocation mechanism, it responsible recovering unused scheduled tasks redistributing remaining budget, which conducive using more expensive/powerful resources accelerate time unscheduled tasks; 2) duplication-based strives selectively predecessors, such advancing trying constraints. At last, carry out four groups experiments, three randomly generated another one actual workflows, compare proposed TDSA baseline algorithms. Experimental confirm overwhelming superiority (up 17.4%) improving utilization 31.6%).