Research On Resource Pricing Mechanism And Trading Strategy Of Grid

With the development of grid computing, the QoS requirements of grid applications have become more and more complex. How to manage the distributed resources and provide satisfying QoS for grid application is always the key issue in grid environments, especially when the capability of grid resources is limited by various factors.In this paper, the popular computing economic models are studied in detail first. Then, the grid resource pricing mechanisms and trading strategies that based on game theory were summarized and compared. Finally, a continuous double auction model for the relaxed reservation of grid resource is proposed. The concept of relaxed time is introduced into the pricing strategy of the model.The buyers determine their price by the amount of remaining time and remaining resources.The sellers ask price under the workload of the system.The bid will increase when the remaining time and remaining resources are reducing, and when the workload of the grid system is reducing the ask price will reduce.The simulation result show that the model proposed is better than the continuous double auction model based on the conventional reservation strategy.The profit of resource provider increases by 21% and the rate of resource utilization increases by 15%.