This is because the Star Student Project clusters and grids normally only serve for certain institutions. The scientists may store the data sets that are most valuable to them, based on the storage capacity of the system. However, the Star Student Project storage capacities are limited in many computing systems, where the scientists have to delete the generated data sets after usage and regenerate them whenever they need to be reused. The data sets storage strategy proposed in this paper is generic. It can be used in any computation- and data intensive applications with different price models of cloud services.
https://www.blogger.com/blogger.g?blogID=598676635414051766#allposts