Performance analysis of computational biology applications on hierarchical Grid systems

The paper studies the performance of computational biology applications on a hierarchical Grid system. These applications are computationally intensive and range from regular patterns (such as database searching applications) to irregular structured patterns (such as building phylogenetic trees). Grid systems are characterized by resources with different computational power and by networks' with widely varying bandwidth. This makes the development or adaptation of computational biology applications on Grid systems challenging. In this paper, we demonstrate how to develop or adapt computational biology applications for hierarchical Grid systems in order to achieve good performance. In addition, we investigate how different bandwidths between computational resources in Grid systems influence the applications' performance.