Implementing atomic sequences on uniprocessors using rollforward

David Mosberger, Peter Druschel, Larry Lee Peterson

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

This article presents a software-only solution to the synchronization problem for uniprocessors. The idea is to execute atomic sequences without any hardware protection, and in the rare case of preemption, to roll the sequence forward to the end, thereby preserving atomicity. One of the proposed implementations protects atomic sequences without any memory-accesses. This is significant as it enables execution at CPU-speeds, rather than memory-speeds. The benefit of this method increases with the frequency at which atomic sequences are executed. It therefore encourages the building of systems with fine-grained synchronization. This has the additional advantage of reducing average latency. Experiments demonstrate that this technique has the potential to outperform even the best hardware mechanisms. The main contribution of this article is to discuss operating-system related issues of rollforward and to demonstrate its practicality, both in terms of flexibility and performance.

Original languageEnglish (US)
Pages (from-to)1-23
Number of pages23
JournalSoftware - Practice and Experience
Volume26
Issue number1
StatePublished - Jan 1996

Fingerprint

Synchronization
Hardware
Data storage equipment
Program processors
Experiments

Keywords

  • Interrupts
  • Locking
  • Rollback

ASJC Scopus subject areas

  • Computer Graphics and Computer-Aided Design
  • Software

Cite this

Implementing atomic sequences on uniprocessors using rollforward. / Mosberger, David; Druschel, Peter; Peterson, Larry Lee.

In: Software - Practice and Experience, Vol. 26, No. 1, 01.1996, p. 1-23.

Research output: Contribution to journalArticle

@article{94488f3e8f154a60b022327aa2eca7ff,
title = "Implementing atomic sequences on uniprocessors using rollforward",
abstract = "This article presents a software-only solution to the synchronization problem for uniprocessors. The idea is to execute atomic sequences without any hardware protection, and in the rare case of preemption, to roll the sequence forward to the end, thereby preserving atomicity. One of the proposed implementations protects atomic sequences without any memory-accesses. This is significant as it enables execution at CPU-speeds, rather than memory-speeds. The benefit of this method increases with the frequency at which atomic sequences are executed. It therefore encourages the building of systems with fine-grained synchronization. This has the additional advantage of reducing average latency. Experiments demonstrate that this technique has the potential to outperform even the best hardware mechanisms. The main contribution of this article is to discuss operating-system related issues of rollforward and to demonstrate its practicality, both in terms of flexibility and performance.",
keywords = "Interrupts, Locking, Rollback",
author = "David Mosberger and Peter Druschel and Peterson, {Larry Lee}",
year = "1996",
month = "1",
language = "English (US)",
volume = "26",
pages = "1--23",
journal = "Software - Practice and Experience",
issn = "0038-0644",
publisher = "John Wiley and Sons Ltd",
number = "1",

}

TY - JOUR

T1 - Implementing atomic sequences on uniprocessors using rollforward

AU - Mosberger, David

AU - Druschel, Peter

AU - Peterson, Larry Lee

PY - 1996/1

Y1 - 1996/1

N2 - This article presents a software-only solution to the synchronization problem for uniprocessors. The idea is to execute atomic sequences without any hardware protection, and in the rare case of preemption, to roll the sequence forward to the end, thereby preserving atomicity. One of the proposed implementations protects atomic sequences without any memory-accesses. This is significant as it enables execution at CPU-speeds, rather than memory-speeds. The benefit of this method increases with the frequency at which atomic sequences are executed. It therefore encourages the building of systems with fine-grained synchronization. This has the additional advantage of reducing average latency. Experiments demonstrate that this technique has the potential to outperform even the best hardware mechanisms. The main contribution of this article is to discuss operating-system related issues of rollforward and to demonstrate its practicality, both in terms of flexibility and performance.

AB - This article presents a software-only solution to the synchronization problem for uniprocessors. The idea is to execute atomic sequences without any hardware protection, and in the rare case of preemption, to roll the sequence forward to the end, thereby preserving atomicity. One of the proposed implementations protects atomic sequences without any memory-accesses. This is significant as it enables execution at CPU-speeds, rather than memory-speeds. The benefit of this method increases with the frequency at which atomic sequences are executed. It therefore encourages the building of systems with fine-grained synchronization. This has the additional advantage of reducing average latency. Experiments demonstrate that this technique has the potential to outperform even the best hardware mechanisms. The main contribution of this article is to discuss operating-system related issues of rollforward and to demonstrate its practicality, both in terms of flexibility and performance.

KW - Interrupts

KW - Locking

KW - Rollback

UR - http://www.scopus.com/inward/record.url?scp=0029754353&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0029754353&partnerID=8YFLogxK

M3 - Article

VL - 26

SP - 1

EP - 23

JO - Software - Practice and Experience

JF - Software - Practice and Experience

SN - 0038-0644

IS - 1

ER -