An integrated high-performance distributed file system based on NFS

In many computer labs, there is large number of computers with unused drive space. There may also be relatively large quantities of data to be stored. The goal of this proposal was to develop a system for storing data distributed over many computers, with enough redundancy so that data can still be recovered if several of the machines are unavailable (due to inevitable hardware failure). A new approach of building a high performance distributed file system, by integrating many NFS services, is presented in this paper. The system is aimed at providing a simple and convenient way to achieve high aggregate I/O bandwidth. The system uses a hyper structure to integrate multiple NFS file systems. It provides multiple data layouts to effectively distribute file data among network nodes. Performance evaluations demonstrate that the system has very good data access bandwidth with near perfect scalability, while still maintains an acceptable meta data throughput.