Naming, Migration, and Replication in NFSv4
نویسندگان
چکیده
The NFSv4 protocol includes features intended to enhance its use in global, wide-area networks. In this paper we describe the design and implementation of a prototype for global name space, transparent migration and fine-grained replication that uses and extends these features. In our system, directory migration and replication use the FS_LOCATIONS attribute to redirect I/O requests. For read/write replication, server redirection provides concurrency and consistency during replica updates. Strict consistency is guaranteed when the system is free of failures. In case of network partition or server failure, we offer the option of increasing availability while relaxing consistency by supporting ordered writes instead of strict consistency. The result is a system built on the Internet standard for distributed filing that features good performance and scalability, high availability, and fine-grained consistency.
منابع مشابه
Naming, Migration, and Replication for NFSv4
In this paper, we discuss a global name space for NFSv4 and mechanisms for transparent migration and replication. By convention, any file or directory name beginning with /nfs on an NFS client is part of this shared global name space. Our system supports file system migration and replication through DNS resolution, provides directory migration and replication using builtin NFSv4 mechanisms, and...
متن کاملThe Object Replication Method for the Efficient Migration of Agent
The migration method of a mobile agent affects the entire performance of the distributed system. Most of the existing migration methods have the slow-witted structures due to the fixed order migration that cause the subsequent accumulation of the task results. They can induce various problems such as the failure of host, network obstacles, service absence, and increased traffic resulting in the...
متن کاملWS-Naming: location migration, replication, and failure transparency support for Web Services
Naming transparencies, i.e., abstracting the name and binding of the entity being used from the endpoints that are actually doing the work, are used in distributed systems to simplify application development by hiding the complexity of the environment. In this paper we demonstrate how to apply traditional distributed systems naming and binding techniques in the Web Services realm. Specifically,...
متن کاملNFSv4 as the Building Block for Fault Tolerant Applications
We propose extensions to the NFSv4 client architecture that provide recovery services, checkpointing and logging to parallel applications. Fault-tolerance relies on NFSv4 clients forming client clusters that share delegation state and transfer data among their local caches. We contend that parallel extensions to NFSv4 should not be limited to file system technologies, such as parallelizing or v...
متن کاملReplication Control in Distributed File Systems
We present a replication control protocol for distributed file systems that can guarantee strict consistency or sequential consistency while imposing no performance overhead for normal reads. The protocol uses a primary-copy scheme with server redirection when concurrent writes occur. It tolerates any number of component omission and performance failures, even when these lead to network partiti...
متن کامل