Abstract
Deadline guarantee is an important QoS requirement for some critical scientific workflow applications. However, the resource heterogeneity and the unpredictable workloads make it difficult for grid system to provide efficient deadline-guarantee service for those applications. Recent IaaS providers, such as Amazon's EC2, provide virtualized on-demand computing resources on a pay-per-use model, which can be aggregated to the existing grid resource pool to enhance deadline-guarantee of scientific workflow. In this paper, a novel workflow scheduling algorithm DGESA is proposed. First, we evaluate the degree of deadline-guarantee for subtasks of workflow in grid system based on proposed probabilistic deadline guarantee model. Then, proper cloud resources are selected as an accelerator to enhance the deadline-guarantee of subtasks. The experimental results show that proposed algorithm achieves better performance than other algorithms on user's deadline guarantee.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
