wide-area cooperative storage with cfs（广域合作存储与慢性疲劳综合症）.pdfVIP

Wide-area cooperative storage with CFS Frank Dabek, M. Frans Kaashoek, David Karger, Robert Morris, Ion Stoica MIT Laboratory for Computer Science chord@ /chord/ Abstract none solves all of them. CFS (the Cooperative File System) is a The Cooperative File System (CFS) is a new peer-to-peer read- new design that meets all of these challenges. only storage system that provides provable guarantees for the ef- A CFS file system exists as a set of blocks distributed over the ficiency, robustness, and load-balance of file storage and retrieval. available CFS servers. CFS client software interprets the stored CFS does this with a completely decentralized architecture that can blocks as file system data and meta-data and presents an ordinary read-only file-system interface to applications. The core of the CFS scale to large systems. CFS servers provide a distributed hash table (DHash) for block storage. CFS clients interpret DHash blocks as software consists of two layers, DHash and Chord. The DHash a file system. DHash distributes and caches blocks at a fine granu- (distributed hash) layer performs block fetches for the client, dis- larity to achieve load balance, uses replication for robustness, and tributes the blocks among the servers, and maintains cached and decreases latency with server selection. DHash finds blocks using replicated copies. DHash uses the Chord [31] distributed lookup