Abstract: Aiming at relaxed reservation of grid resource, an improved continuous double auction model is proposed. The concept of relaxed time is applied into the pricing strategy model, in which the buyers determine their price according to the amount of remaining time and remaining resources, and the sellers ask price considering the workload of the system. Simulation experimental results show that the model can increase the profit of resource provider by 21% and the rate of resource utilization by 15%.

Key words: grid computing, advanced reservation, relaxed reservation, continuous double auction, resource utilization rate

摘要： 针对网格资源的松弛预留问题，提出一种改进的连续双向拍卖模型。在网格资源定价策略中，引入松弛时间保证更高的任务预留接纳率。在用户出价和要价策略中，买方通过剩余时间和剩余资源量出价，卖方根据负载情况要价。仿真实验结果表明，对于具有费用约束的网格任务，该模型能增加约21%的资源总收益，提高约15%的资源利用率。

关键词: 网格计算, 提前预留, 松弛预留, 连续双向拍卖, 资源利用率

CLC Number: 

• TP393