Job allocation and scheduling for data transfer is a fundamental issue for achieving high performance in data grid environments. In this paper, we propose a new algorithm that combines job allocation with scheduling dynamically based on resource quality. The algorithm takes resource failure into consideration and provides a re-allocation mechanism, so it can utilize limited amounts of resources efficiently and enhance the reliability of data transfer in data grid. A definition of resource quality is given in the paper as well, which consists of information about CPU and bandwidth of the grid storage node that resource resides. To reflect historical performance of resource, a new instance of ant algorithm is designed for calculating and updating this resource quality. Based on this quality, the job allocation and scheduling algorithm can take full advantage of the high performance resources and balance the load among resources at the same time. Experimental results show that the algorithm satisfies the expectations
展开▼
