Data mining applications are composed of computing-intensive processing tasks, which are natural candidates for execution on high performance, high throughput platforms such as PC clusters and computational grids. Besides, some data-mining algorithms can be implemented as Bag-of-Tasks (BoT) applications, which are composed of parallel, independent tasks. Due to its own nature, the adaptation of BoT applications for the grid is straightforward. In this sense, this work proposes a scheduling algorithm for running BoT data mining applications on grid platforms. The proposed algorithm is evaluated by means of several experiments, and the obtained results show that it improves both scalability and performance of such applications.
展开▼