Farhad Ghorbanzadeh - Department of Computer Engineering, Tabriz Branch, Islamic Azad University, Tabriz, Iran
Ali Asghar Pourhaji Kazem - Department of Computer Engineering, Tabriz Branch, Islamic Azad University Tabriz, Iran

In grid computing environments, users have access to multiple resources that may be distributed geographically. Thus, resource allocation and scheduling is an important procedure in achieving high performance on grid computing. For this, market-based method is suitable because both resource consumer and resource provider have enough incentive to stay and play in the grid. One type of market-based is auction model that continuous double auction (CDA) is a kind of that. This method has already studied and results showed that it has a good performance in successful execution and resource utilization rates. In this paper we want to combine CDA with artificial neural network to improve the performance of this method. The simulation results show that our proposed method decrease the jobs failure rate.

Grid, Resource Allocation, Neural Network