GridFS: Targeting Data Sharing in Grid Environments

Abstract

GridFS is a system that enables data sharing in a cluster or grid environment. By deploying a set of servers over several nodes, it is possible to build a federated wide area file system integrating tera scale sized data. The data is stored in different hosts under a single name space controlled by GridFS. It was designed and developed considering interoperability, scalability and performance issues. Also, GridFS provides special functions for process schedulers, such as file transfer rate estimates, and supports legacy applications