On the resource allocation and pricing strategies in Compute Clouds using bargaining approaches

In this paper, we consider addressing the resource allocation and pricing strategies in a Compute Cloud for both independent tasks and tasks from workflow schemes. Workflow scheduling of tasks is an important problem due to the fact that individual sub-tasks constituting the workflow may demand additional resources and hence may stall the entire process. We employ two axiomatic bargaining approaches (Nash Bargaining Solution (NBS) and Raiffa Bargaining Solution (RBS)) proposed in the literature to formulate the problem and derive an optimal solution for allocating virtual CPU instances in a Compute Cloud for both independent tasks and workflow tasks. We also analyze the effectiveness of our strategies via rigorous simulation experiments and we show that our strategies are adaptable to the requirements by the Cloud service providers (CSPs) in estimating the resource requirements. Further, we show that NBS ensures proportional fairness whereas RBS can handle real-time task arrivals and task dynamics. Finally we introduce the concept of asymmetric pricing scheme in which a user can specify his budget constraints and CSPs can attempt to maximize the revenue without sacrificing the performance. This asymmetric bargaining approach is an important contribution in this work which allows the CSP to choose different parameters such as deadline and/or budget requirements for deriving optimal resource allocation. The deadline based resource allocation is particularly useful for workflow-based applications which have tasks waiting for the completion of other tasks.