A node selection mechanism based on the node usage pattern on campus grid

"Campus grid" consists of a collection of computing resources distributed over a university-wide network. To use computing resources on campus grid efficiently, we propose the node selection mechanism based on the usage pattern of computational nodes in campus grid. In our node selection mechanism, we represent the usage pattern of a computational node by using both the average and the standard deviation of the load every one hour and define a node selection algorithm. To evaluate our proposed mechanism, we use NAS parallel benchmarks. The result shows the effectiveness of our proposed mechanism.