Threshold - Based Negotiation Framework for Grid Resource Allocation

What is it about?

Computational Grid provides virtual powerful computer for solving any large-scale scientific, commercial, and engineering problems by aggregating and sharing heterogeneous resources. However, designing an effective resource allocation method is a complicated task. To undertake the task, several economic-based resource management models have been studied.

Featured Image

Why is it important?

Although bargaining (negotiation) is an economic-based resource allocation method, it has two main defects: i) the possibility of losing a deal is high if trading partners have grasping characteristics meaning that consumers take risk for cheaper prices while providers take risk for more expensive prices as much as possible, ii) the model requires high communication demand when there are a large number of grid entities in the grid. This paper proposes a negotiation framework with a threshold mechanism to cope with the mentioned flaws. Instead of repeating negotiation for every single grid job, the proposed framework requires negotiating only when market conditions adequately change. Simulation results show that a negotiation method is more successful, demands less communication and spends less time on negotiation when it is executed with the proposed framework.