Wide-area cooperative storage with CFS

Proceedings of the eighteenth ACM symposium on Operating systems principles Pub Date : 2001-10-21 DOI:10.1145/502034.502054

F. Dabek, M. Kaashoek, David R Karger, R. Morris, I. Stoica

{"title":"Wide-area cooperative storage with CFS","authors":"F. Dabek, M. Kaashoek, David R Karger, R. Morris, I. Stoica","doi":"10.1145/502034.502054","DOIUrl":null,"url":null,"abstract":"The Cooperative File System (CFS) is a new peer-to-peer read-only storage system that provides provable guarantees for the efficiency, robustness, and load-balance of file storage and retrieval. CFS does this with a completely decentralized architecture that can scale to large systems. CFS servers provide a distributed hash table (DHash) for block storage. CFS clients interpret DHash blocks as a file system. DHash distributes and caches blocks at a fine granularity to achieve load balance, uses replication for robustness, and decreases latency with server selection. DHash finds blocks using the Chord location protocol, which operates in time logarithmic in the number of servers.CFS is implemented using the SFS file system toolkit and runs on Linux, OpenBSD, and FreeBSD. Experience on a globally deployed prototype shows that CFS delivers data to clients as fast as FTP. Controlled tests show that CFS is scalable: with 4,096 servers, looking up a block of data involves contacting only seven servers. The tests also demonstrate nearly perfect robustness and unimpaired performance even when as many as half the servers fail.","PeriodicalId":263344,"journal":{"name":"Proceedings of the eighteenth ACM symposium on Operating systems principles","volume":"58 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2001-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1869","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings of the eighteenth ACM symposium on Operating systems principles","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/502034.502054","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 1869

Abstract

The Cooperative File System (CFS) is a new peer-to-peer read-only storage system that provides provable guarantees for the efficiency, robustness, and load-balance of file storage and retrieval. CFS does this with a completely decentralized architecture that can scale to large systems. CFS servers provide a distributed hash table (DHash) for block storage. CFS clients interpret DHash blocks as a file system. DHash distributes and caches blocks at a fine granularity to achieve load balance, uses replication for robustness, and decreases latency with server selection. DHash finds blocks using the Chord location protocol, which operates in time logarithmic in the number of servers.CFS is implemented using the SFS file system toolkit and runs on Linux, OpenBSD, and FreeBSD. Experience on a globally deployed prototype shows that CFS delivers data to clients as fast as FTP. Controlled tests show that CFS is scalable: with 4,096 servers, looking up a block of data involves contacting only seven servers. The tests also demonstrate nearly perfect robustness and unimpaired performance even when as many as half the servers fail.

查看原文

微信好友朋友圈 QQ好友复制链接

本刊更多论文

与粮安委广域合作仓储

协作文件系统(Cooperative File System, CFS)是一种新型的点对点只读存储系统，它为文件存储和检索的效率、健壮性和负载均衡提供了可证明的保证。CFS通过一个完全分散的架构来实现这一点，该架构可以扩展到大型系统。CFS服务器为块存储提供分布式哈希表(DHash)。CFS客户端将dash块解释为文件系统。DHash以细粒度分发和缓存块以实现负载平衡，使用复制实现鲁棒性，并通过服务器选择减少延迟。DHash使用Chord定位协议查找区块，该协议在服务器数量上按对数时间进行操作。CFS是使用SFS文件系统工具包实现的，可以在Linux、OpenBSD和FreeBSD上运行。在全局部署的原型上的经验表明，CFS向客户机交付数据的速度与FTP一样快。受控测试表明CFS是可扩展的:使用4,096台服务器，查找数据块只需要联系7台服务器。测试还展示了近乎完美的健壮性和不受影响的性能，即使多达一半的服务器出现故障。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文去求助

来源期刊

Proceedings of the eighteenth ACM symposium on Operating systems principles

自引率

0.00%

发文量

期刊最新文献

An empirical study of operating systems errors Building a robust software-based router using network processors BASE: using abstraction to improve fault tolerance Information and control in gray-box systems The costs and limits of availability for replicated services