Diskless checkpointing

Author

Plank, James S. ; Li, Kai ; Puening, Michael A.

Author_Institution

Dept. of Comput. Sci., Tennessee Univ., Knoxville, TN, USA

Volume

9

Issue

10

fYear

1998

fDate

10/1/1998 12:00:00 AM

Firstpage

972

Lastpage

986

Abstract

Diskless Checkpointing is a technique for checkpointing the state of a long-running computation on a distributed system without relying on stable storage. As such, it eliminates the performance bottleneck of traditional checkpointing on distributed systems. In this paper, we motivate diskless checkpointing and present the basic diskless checkpointing scheme along with several variants for improved performance. The performance of the basic scheme and its variants is evaluated on a high-performance network of workstations and compared to traditional disk-based checkpointing. We conclude that diskless checkpointing is a desirable alternative to disk-based checkpointing that can improve the performance of distributed applications in the face of failures

Keywords

distributed processing; fault tolerant computing; system recovery; disk-based checkpointing; diskless checkpointing; distributed applications; distributed system; high-performance network; long-running computation; performance bottleneck; Checkpointing; Computer Society; Distributed computing; Error correction codes; Fault tolerance; Fault tolerant systems; Hardware; Programming environments; Redundancy; Workstations;

fLanguage

English

Journal_Title

Parallel and Distributed Systems, IEEE Transactions on

Publisher

ieee

ISSN

1045-9219

Type

jour

DOI

10.1109/71.730527

Filename

730527