Scheduling Scientific Workflow Based Application Using ACO in Public Cloud

Scientific workflows comprising of many computational tasks including data dependency may require multiple and heterogeneous amount of computing resources during runtime. Scheduling such workflows with the objective of achieving minimal makespan and cost and maximal resource usage is a challenge in any computing environment. The researchers aim at developing novel algorithm to schedule scientific workflow in an emerging computing area of public cloud where they can avail mass heterogeneous amount of resources on pay-per-use mode. The proposed algorithm uses Ant Colony Optimization (ACO) approach to optimize the scheduling strategy in order to achieve the objective of minimal makespan. The researchers have compared the results with other popular algorithm Heterogeneous Earliest Finish Time (HEFT).The experimental results show that the proposed algorithm has significant potential to achieve the objective. Keyword Task Scheduling, Ant Colony Optimization, Cloud Computing, Scientific Workflow.