Wide-area cooperative storage with CFS

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.

Making Chord Robust to Byzantine Attacks

“…As in [2,3,7,9], we make use of the concept of small sets of peers working together as a single functional unit. Central to our protocol is the notion of a swarm 7 .…”

Overlay networks a scalable alternative for p2p

“…Researchers are also successfully deploying overlay networks as part of distributed storage systems, such as the cooperative file system. 7 CFS interprets the Chord network's stored values as a file system, and includes features such as replication for increased robustness. Finally, researchers have applied the Pastry system to various end-user applications, such as cooperative Web caching, group notification, and instant messaging.…”

Brocade: Landmark Routing on Overlay Networks

“…In related work, the Cooperative File System [2] leverages nodes with more resources by allowing them to host additional virtual nodes in the system, each representing one quantum of resource. This quantification is directed mostly at storage requirements, and CFS does not propose a mechanism for exploiting network topology knowledge.…”

“…These architectures make use of name-based routing to route requests for objects or files to a nearby replica. Applications built on such systems ( [2], [3], [7]), depend on reliable and fast message routing to a destination node, given some unique identifier.…”