Sharing memory with semi-Byzantine clients and faulty storage servers

Author

Attiya, Hagit ; Bar-Or, Amir

Author_Institution

Dept. of Comput. Sci., Technion, Haifa, Israel

fYear

2003

fDate

6-18 Oct. 2003

Firstpage

371

Lastpage

378

Abstract

This paper presents several fault-tolerant simulations of a single-writer multi-reader regular register in storage systems. One simulation tolerates fail-stop failures of storage servers and require a majority of nonfaulty servers, while the other simulation tolerates Byzantine failures and requires that two-thirds of the servers to be nonfaulty. A construction of Afek et al.(1995) is used to mask semi-Byzantine failures of clients that result in erroneous write operations. The simulations are used to derive Paxos algorithms that tolerate semi-Byzantine failures of clients as well as failstop or Byzantine failures of storage servers.

Keywords

client-server systems; fault simulation; fault tolerant computing; redundancy; shared memory systems; storage area networks; Byzantine failure; Paxos algorithms; erroneous write operations; fail-stop failures; fault-tolerant simulations; faulty storage servers; memory sharing; semiByzantine clients; single-writer multireader register; Computational modeling; Computer science; Computer simulation; Delay; Fault tolerance; Fault tolerant systems; File servers; Laboratories; Redundancy; Registers;

fLanguage

English

Publisher

ieee

Conference_Titel

Reliable Distributed Systems, 2003. Proceedings. 22nd International Symposium on

ISSN

1060-9857

Print_ISBN

0-7695-1955-5

Type

conf

DOI

10.1109/RELDIS.2003.1238090

Filename

1238090