A Throughput Maximisation Strategy for Scheduling Transaction Intensive Workflows on SwinDeW-G

With the rapid development of e-business, workflow systems now have to deal with transaction intensive workflows whose main characteristic is the huge number of concurrent workflow instances. For such workflows, it is important to maximise the overall throughput to provide good quality of service. However, most existing scheduling algorithms are designed for scheduling of a single complex scientific workflow instance and are not efficient enough for scheduling transaction intensive workflows. To address this problem, we propose a throughput maximisation strategy (TMS) which contains two specific algorithms for scheduling transaction-intensive workflows at instance and task levels, respectively. The first algorithm called Opposite Average Load algorithm tries to maximise the overall throughput by pursuing overall load balance at instance level, whereas the second algorithm called Extended Min-Min algorithm dedicates to further maximise the overall throughput at task level by increasing resource utilization rate within each local autonomous group. The comparison and simulation performed on Swinburne Decentralised Workflow for Grid (SwinDeW-G), a p2p based grid workflow environment, demonstrate that our strategy can improve the overall throughput significantly over existing scheduling algorithms when scheduling transaction intensive workflows. Copyright © 2008 John Wiley & Sons, Ltd.