A survey of rollback-recovery protocols in message-passing systems

by: (mootaz) EN Elnozahy, Lorenzo Alvisi, Yi-Min Wang, David B Johnson

ACM Comput. Surv., Vol. 34, No. 3. (September 2002), pp. 375-408.

View FullText article

ACM, DOI

Reviews [Write a review of this article]

There are no reviews of this article

Notes for this article

Gnurou a 1 notes publique et 0 note privées sur cette article . If you are Gnurou then you can log in to see the private publique .

Find related articles from these CiteULike users

kistijantoro, damaru, Gnurou

Find related articles with these CiteULike tags

no-tag, piecewise_deterministic, recovery, relatedwork

Résumé

This survey covers rollback-recovery techniques that do not require special language constructs. In the first part of the survey we classify rollback-recovery protocols into checkpoint-based and log-based. Checkpoint-based protocols rely solely on checkpointing for system state restoration. Checkpointing can be coordinated, uncoordinated, or communication-induced. Log-based protocols combine checkpointing with logging of nondeterministic events, encoded in tuples called determinants. Depending on how determinants are logged, log-based protocols can be pessimistic, optimistic, or causal. Throughout the survey, we highlight the research issues that are at the core of rollback-recovery and present the solutions that currently address them. We also compare the performance of different rollback-recovery protocols with respect to a series of desirable properties and discuss the issues that arise in the practical implementations of these protocols.

@article{citeulike:364695, abstract = {This survey covers rollback-recovery techniques that do not require special language constructs. In the first part of the survey we classify rollback-recovery protocols into checkpoint-based and log-based. Checkpoint-based protocols rely solely on checkpointing for system state restoration. Checkpointing can be coordinated, uncoordinated, or communication-induced. Log-based protocols combine checkpointing with logging of nondeterministic events, encoded in tuples called determinants. Depending on how determinants are logged, log-based protocols can be pessimistic, optimistic, or causal. Throughout the survey, we highlight the research issues that are at the core of rollback-recovery and present the solutions that currently address them. We also compare the performance of different rollback-recovery protocols with respect to a series of desirable properties and discuss the issues that arise in the practical implementations of these protocols.}, address = {New York, NY, USA}, author = {Elnozahy, (mootaz) E. N. and Alvisi, Lorenzo and Wang, Yi-Min and Johnson, David B. }, citeulike-article-id = {364695}, doi = {10.1145/568522.568525}, issn = {0360-0300}, journal = {ACM Comput. Surv.}, keywords = {recovery, relatedwork}, month = {September}, number = {3}, pages = {375--408}, posted-at = {2007-12-14 08:04:32}, priority = {3}, publisher = {ACM Press}, title = {A survey of rollback-recovery protocols in message-passing systems}, url = {http://portal.acm.org/citation.cfm?id=568525}, volume = {34}, year = {2002} }

RIS record

TY - JOUR ID - citeulike:364695 L3 - citeulike-article-id:364695 TI - A survey of rollback-recovery protocols in message-passing systems JF - ACM Comput. Surv. VL - 34 IS - 3 SP - 375 EP - 408 AD - New York, NY, USA PB - ACM Press SN - 0360-0300 N2 - This survey covers rollback-recovery techniques that do not require special language constructs. In the first part of the survey we classify rollback-recovery protocols into checkpoint-based and log-based. Checkpoint-based protocols rely solely on checkpointing for system state restoration. Checkpointing can be coordinated, uncoordinated, or communication-induced. Log-based protocols combine checkpointing with logging of nondeterministic events, encoded in tuples called determinants. Depending on how determinants are logged, log-based protocols can be pessimistic, optimistic, or causal. Throughout the survey, we highlight the research issues that are at the core of rollback-recovery and present the solutions that currently address them. We also compare the performance of different rollback-recovery protocols with respect to a series of desirable properties and discuss the issues that arise in the practical implementations of these protocols. KW - recovery KW - relatedwork AU - Elnozahy, (mootaz) AU - Alvisi, Lorenzo AU - Wang, Yi-Min AU - Johnson, David PY - 2002/09// UR - http://portal.acm.org/citation.cfm?id=568525 ER -

RIS BibTeX