Diskless checkpointing

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

doi:10.1109/71.730527

Diskless checkpointing

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

Research output: Contribution to journal › Article › peer-review

266 Scopus citations

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.

Original language	English (US)
Pages (from-to)	972-986
Number of pages	15
Journal	IEEE Transactions on Parallel and Distributed Systems
Volume	9
Issue number	10
DOIs	https://doi.org/10.1109/71.730527
State	Published - 1998

All Science Journal Classification (ASJC) codes

Signal Processing
Hardware and Architecture
Computational Theory and Mathematics

Keywords

Checkpointing
Copy-on-write
Error-correcting codes
Fault tolerance
Memory redundancy
RAID systems
Rollback recovery

Access to Document

10.1109/71.730527

Cite this

@article{7fc2ea9cadd94d1a8ac94a62e7f4cf19,

title = "Diskless checkpointing",

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 = "Checkpointing, Copy-on-write, Error-correcting codes, Fault tolerance, Memory redundancy, RAID systems, Rollback recovery",

author = "Plank, {James S.} and Kai Li and Puening, {Michael A.}",

note = "Funding Information: This material is based upon work supported by the U.S. National Science Foundation under grants CCR-9409496, CDA-9529459, and CCR-9703390, and by the ORAU Junior Faculty Achievement Award. The authors thank Nitin Vaidya and the anonymous referees for their helpful comments on this work.",

year = "1998",

doi = "10.1109/71.730527",

language = "English (US)",

volume = "9",

pages = "972--986",

journal = "IEEE Transactions on Parallel and Distributed Systems",

issn = "1045-9219",

publisher = "IEEE Computer Society",

number = "10",

}

TY - JOUR

T1 - Diskless checkpointing

AU - Plank, James S.

AU - Li, Kai

AU - Puening, Michael A.

N1 - Funding Information: This material is based upon work supported by the U.S. National Science Foundation under grants CCR-9409496, CDA-9529459, and CCR-9703390, and by the ORAU Junior Faculty Achievement Award. The authors thank Nitin Vaidya and the anonymous referees for their helpful comments on this work.

PY - 1998

Y1 - 1998

N2 - 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.

AB - 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.

KW - Checkpointing

KW - Copy-on-write

KW - Error-correcting codes

KW - Fault tolerance

KW - Memory redundancy

KW - RAID systems

KW - Rollback recovery

UR - http://www.scopus.com/inward/record.url?scp=0032179680&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0032179680&partnerID=8YFLogxK

U2 - 10.1109/71.730527

DO - 10.1109/71.730527

M3 - Article

AN - SCOPUS:0032179680

SN - 1045-9219

VL - 9

SP - 972

EP - 986

JO - IEEE Transactions on Parallel and Distributed Systems

JF - IEEE Transactions on Parallel and Distributed Systems

IS - 10

ER -

Diskless checkpointing

Abstract

All Science Journal Classification (ASJC) codes

Keywords

Access to Document

Other files and links

Fingerprint

Cite this