Name space consistency in the Pangaea wide - area file system HPL
نویسندگان
چکیده
Pangaea is a wide-area file system that enables ad-hoc collaboration in multi-national corporations or in distributed groups of users. This paper describes Pangaea’s approach for keeping the file-system’s name space consistent and proves its correctness. Maintaining the name space is a simple matter in traditional file systems that store the entire volume in a single node. It is not so in Pangaea, because of the two key techniques it employs to improve performance and availability in a wide area— pervasive replication that lets each file be replicated on its own set of nodes on demand from users, and optimistic replication that lets updates be issued on any replicas at any time. A naive implementation may leave some files unreachable in the name space or some directory entries pointing to non-existent files. To detect conflicting updates and inform all affected replicas about the resolution outcome reliably, Pangaea embeds, in each file, a data structure called backpointer that authoritatively defines the file’s location in the file-system’s name space. Conflicting directory operations are detected by a replica of the (child) file as a discrepancy in the value of the backpointer. The replica can then unilaterally resolve conflicts and disseminate the conflict resolution outcome to the the parent directories.
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 کاملDataStations: Ubiquitous Transient Storage for Mobile Users
In this paper, we describe DataStations, an architecture that provides ubiquitous transient storage to arbitrary mobile applications. Mobile users can utilize a nearby DataStation as a proxy cache for their remote home file servers, as a file server to meet transient storage needs, and as a platform to share data and collaborate with other users over the wide area. A user can roam among DataSta...
متن کاملAvailability and Consistency Tradeoffs in the Echo Distributed File System
Workstations typically depend on remote servers accessed over a network for such services as mail, printing, storing files, booting, and time. The availability of these remote services has a major impact on the usability of the workstation. Availability can be increased by replicating the servers. In the Echo distributed file system at DEC SRC, two different replication techniques are employed,...
متن کاملExplicit Control in a Batch-Aware Distributed File System
We present the design, implementation, and evaluation of the Batch-Aware Distributed File System (BAD-FS), a system designed to orchestrate large, I/O-intensive batch workloads on remote computing clusters distributed across the wide area. BAD-FS consists of two novel components: a storage layer that exposes control of traditionally fixed policies such as caching, consistency, and replication; ...
متن کاملExplicit Control in the Batch-Aware Distributed File System
We present the design, implementation, and evaluation of the Batch-Aware Distributed File System (BAD-FS), a system designed to orchestrate large, I/O-intensive batch workloads on remote computing clusters distributed across the wide area. BAD-FS consists of two novel components: a storage layer that exposes control of traditionally fixed policies such as caching, consistency, and replication; ...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2002