Searches Based On The Network Of Distributed File System

Along with on line data fast growth, the traditional server becomes the network service bottleneck. The grid technology is playing the huge role in the high performance memory area of technology. However, Data Grid, the data management solution for WAN, is unable to satisfy the mass memory request for the magnanimous user. Therefore, network-based distributed file system validity has further demonstrated out.NFS is the earliest file system realizing network document sharing. Non-keeping system among state information is an important characteristics of NFS. It mainly applies on the small LAN. AFS then can expand to several thousand user's distributed file system. It is the first file system that uses the permanent buffer strategy in the client side. But AFS has not solved distributional data or other documents required large random, also has not processed the inconsistent of the data. DFS is an item of file service which Windows 2000(Server version)/2003 system. In the client side, DFS has carried on the transformation to AFS, the main difference lies in the use of outstanding document pellet to share. While Coda is the first distributed file system which supports disconnected operation. The ancestor of Coda is the AFS distributed file system, it is suitable for the environment in which the network connection is not stable.Coda does not effectively tell user what is wrong with it. The author present the improvement frame in this foundation——an overview of graphical interface called the CodaConsole,that makes caching translucent to Coda users in the presence of disconnected or weakly connected operation. Hence the availability of Coda is improved greatly. The user also can understand system behavior well. At the same time, the author has produced the concrete realization plan and the improvement effect.