Low-Latency, Concurrent Checkpointing for Parallel Programs

Kai Li

doi:10.1109/71.298215

Low-Latency, Concurrent Checkpointing for Parallel Programs

Kai Li

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

87 Scopus citations

Abstract

This short note presents the results of an implementation of several algorithms for checkpointing and restarting parallel programs on shared-memory multiprocessors. The algorithms are compared according to the metrics of overall checkpointing time, overhead imposed by the checkpointer on the target program, and amount of time during which the checkpointer interrupts the target program. The best algorithm measured achieves its efficiency through a variation of copy-on-write, which allows the most time-consuming operations of the checkpoint to be overlapped with the running of the program being checkpointed.

Original language	English (US)
Pages (from-to)	874-879
Number of pages	6
Journal	IEEE Transactions on Parallel and Distributed Systems
Volume	5
Issue number	8
DOIs	https://doi.org/10.1109/71.298215
State	Published - Aug 1994

All Science Journal Classification (ASJC) codes

Signal Processing
Hardware and Architecture
Computational Theory and Mathematics

Keywords

Checkpointing fault tolerance copy-on-write multiprocessing backward error recovery
Index Terms—

Access to Document

10.1109/71.298215

Cite this

@article{1f2d4b6eaa8346e1a1d4eb552c3c5d92,

title = "Low-Latency, Concurrent Checkpointing for Parallel Programs",

abstract = "This short note presents the results of an implementation of several algorithms for checkpointing and restarting parallel programs on shared-memory multiprocessors. The algorithms are compared according to the metrics of overall checkpointing time, overhead imposed by the checkpointer on the target program, and amount of time during which the checkpointer interrupts the target program. The best algorithm measured achieves its efficiency through a variation of copy-on-write, which allows the most time-consuming operations of the checkpoint to be overlapped with the running of the program being checkpointed.",

keywords = "Checkpointing fault tolerance copy-on-write multiprocessing backward error recovery, Index Terms—",

author = "Kai Li",

note = "Funding Information: Manuscript received July 7, 1992; revised July 9, 1993. This work was supported in part by the National Science Foundation under Grants CCR-8814265 and IRI-8909795, and in part by the Digital Equipment External Research Program and Systems Research Center. K. Li is with the Department of Computer Science, Princeton University, Princeton, NJ 08544 USA; e-mail: li@princeton.edu. J. Naughton is with the Department of Computer Science, University of Wisconsin, Madison, WI 53706 USA. J. Plank is with the Department of Computer Science, University of Tennessee, Knoxville, TN 37966 USA. IEEE Log Number 9401208.",

year = "1994",

month = aug,

doi = "10.1109/71.298215",

language = "English (US)",

volume = "5",

pages = "874--879",

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

issn = "1045-9219",

publisher = "IEEE Computer Society",

number = "8",

}

TY - JOUR

T1 - Low-Latency, Concurrent Checkpointing for Parallel Programs

AU - Li, Kai

N1 - Funding Information: Manuscript received July 7, 1992; revised July 9, 1993. This work was supported in part by the National Science Foundation under Grants CCR-8814265 and IRI-8909795, and in part by the Digital Equipment External Research Program and Systems Research Center. K. Li is with the Department of Computer Science, Princeton University, Princeton, NJ 08544 USA; e-mail: li@princeton.edu. J. Naughton is with the Department of Computer Science, University of Wisconsin, Madison, WI 53706 USA. J. Plank is with the Department of Computer Science, University of Tennessee, Knoxville, TN 37966 USA. IEEE Log Number 9401208.

PY - 1994/8

Y1 - 1994/8

N2 - This short note presents the results of an implementation of several algorithms for checkpointing and restarting parallel programs on shared-memory multiprocessors. The algorithms are compared according to the metrics of overall checkpointing time, overhead imposed by the checkpointer on the target program, and amount of time during which the checkpointer interrupts the target program. The best algorithm measured achieves its efficiency through a variation of copy-on-write, which allows the most time-consuming operations of the checkpoint to be overlapped with the running of the program being checkpointed.

AB - This short note presents the results of an implementation of several algorithms for checkpointing and restarting parallel programs on shared-memory multiprocessors. The algorithms are compared according to the metrics of overall checkpointing time, overhead imposed by the checkpointer on the target program, and amount of time during which the checkpointer interrupts the target program. The best algorithm measured achieves its efficiency through a variation of copy-on-write, which allows the most time-consuming operations of the checkpoint to be overlapped with the running of the program being checkpointed.

KW - Checkpointing fault tolerance copy-on-write multiprocessing backward error recovery

KW - Index Terms—

UR - http://www.scopus.com/inward/record.url?scp=0028485392&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0028485392&partnerID=8YFLogxK

U2 - 10.1109/71.298215

DO - 10.1109/71.298215

M3 - Article

AN - SCOPUS:0028485392

SN - 1045-9219

VL - 5

SP - 874

EP - 879

JO - IEEE Transactions on Parallel and Distributed Systems

JF - IEEE Transactions on Parallel and Distributed Systems

IS - 8

ER -

Low-Latency, Concurrent Checkpointing for Parallel Programs

Abstract

All Science Journal Classification (ASJC) codes

Keywords

Access to Document

Other files and links

Fingerprint

Cite this