Data intensive task is becoming one of the most important applications in grid environment. The scale of data sets has been hundreds of terabytes and soon will be petabytes. The primary problem we face is how to organize the geographical distributed storage devices to support the collaborative operations on data in those resources. On the one hand, performance is critical to such application, but on the other hand the diverse network conditions prevent users from getting the same service quality. This paper focuses on how to resolve the above problem, and presents a 2-layered metadata service model in grid environment which utilizes the special locality of users’ distribution and provides a platform for grid data management. We have implemented that 2-layered metadata service model in
ChinaGrid Supporting Platform
(CGSP) – the grid middleware for ChinaGrid project.
Bitte loggen Sie sich ein, um Zugang zu diesem Inhalt zu erhalten