Publication: Reliable and Fault-Tolerant Peer-to-Peer Block Storage
Open/View Files
Date
2002
Published Version
Published Version
Journal Title
Journal ISSN
Volume Title
Publisher
The Harvard community has made this article openly available. Please share how this access benefits you.
Citation
Stein, C.A. (Lex), Michael J. Tucker, and Margo I. Seltzer. 2002. Reliable and Fault-Tolerant Peer-to-Peer Block Storage. Harvard Computer Science Group Technical Report TR-04-02.
Research Data
Abstract
The Charles is a scalable, fault-tolerant, persistent block storage service built from a constantly changing network of potentially faulty or malicious computing nodes. The Charles is designed for a P2P environment: networks characterized by frequent process joins and departures, as well as arbitrary or Byzantine process behavior. The Charles consists of two levels. The top level provides a scalable lookup service. This level is known as the Surface level and is structured as a ring of slowly changing process groups. The bottom level, known as the Submerged level, combines individual processes together into fault-tolerant groups absorbing network dynamics and process failure; whether it is arbitrary behavior or crash failure. The bottom level serves to provide the top level with predictable and well-behaved process groups. Both levels are self-organizing and self-managing. The Charles has no centralized control. The Charles combines block lookup with storage. Fault-tolerance is provided through replication within process groups. Replication is maintained internally without any application state or callbacks.
Description
Other Available Sources
Keywords
Terms of Use
This article is made available under the terms and conditions applicable to Other Posted Material (LAA), as set forth at Terms of Service