The advent of Cloud computing has provided a promising methodology for usage of distributed resources for complex scientific workflow applications. Due to the unique features of cloud technology, such as the pay-as-you-go pricing model and scaling, efficient workflow scheduling is a critical research topic. While most workflow scheduling algorithms are proposed to minimize the overall execution time, cost-driven public cloud services have made cost minimization an emerging and critical issue. Therefore, the objective of this work is to solve the cost optimization problem for scheduling workflows on a commercial Cloud while considering a userdefined deadline constraint. In this paper, a heuristic algorithm for scheduling deadline-constrained workflows is presented - Deadline Constrained Level Based (DCLB)- which uses Level Load Balancing to refine deadline distribution as well as attaining lower communication cost in order to reach the algorithm's goals. Experimental results demonstrate that DCLB compared to existing algorithms, achieves higher cost efficiencies when workflow deadline is met.

Keywords