FADI: a fault tolerant environment for open distributed computing

Author

Osman, T. ; Bargiela, A.

Author_Institution

Dept. of Comput., Nottingham Trent Univ., UK

Volume

147

Issue

3

fYear

2000

fDate

6/1/2000 12:00:00 AM

Firstpage

91

Lastpage

99

Abstract

FADI (fault tolerant distributed environment) is a complete programming environment for the reliable execution of distributed application programs. FADI encompasses all aspects of modern fault-tolerant distributed computing. The built-in user-transparent error detection mechanism covers processor node crashes and hardware transient failures. The mechanism also integrates user-assisted error checks into the system failure model. The nucleus non-blocking checkpointing mechanism combined with a novel selective message logging technique delivers an efficient, low-overhead backup and recovery mechanism for distributed processes. FADI also provides a means of remote automatic process allocation on distributed system nodes

Keywords

distributed programming; open systems; programming environments; software fault tolerance; system recovery; FADI; checkpointing mechanism; distributed application programs; distributed processes; distributed system nodes; fault tolerant distributed environment; fault tolerant environment; fault-tolerant distributed computing; hardware transient failures; low-overhead backup; open distributed computing; processor node crashes; programming environment; recovery mechanism; remote automatic process allocation; selective message logging technique; system failure model; user-assisted error checks; user-transparent error detection;

fLanguage

English

Journal_Title

Software, IEE Proceedings -

Publisher

iet

ISSN

1462-5970

Type

jour

DOI

10.1049/ip-sen:20000702

Filename

888328