Advertisement

Annales Des Télécommunications

, Volume 49, Issue 7–8, pp 460–469 | Cite as

Déterminer un état global dans un système réparti

  • Jean-Michel Hélary
  • Achour Mostefaoui
  • Michel Raynal
Article
  • 21 Downloads

Résumé

Le calcul d’un état global réparti est l’un des paradigmes des problèmes de contrôle des messages entre processus répartis. Après avoir exposé le problème, cet article analyse et présente plusieurs solutions. Celles-ci se distinguent les unes des autres par les hypothèses qu’elles font sur les canaux de communication et les messages de contrôle utilisés. En plus de son caractère synthétique sur le calcul d’un état global réparti, cet article constitue une illustration des hypothèses et difficultés propres au contexte réparti.

Mots clés

Traitement réparti Méthode globale Traitement parallèle Article synthèse Causalité 

Determining a global state in a distributed system

Abstract

Recording a global snapshot of a distributed system is one of the paradigms of distributed control problems. This paper presents and analyses several solutions. They differ by the assumptions made on the communication channels and the use of control messages. Moreover, this paper throws light on assumptions and difficulties inherent to a distributed environment.

Key words

Distributed processing Global method Parallel processing Review Causality 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Bibliographie

  1. [1]
    Acharya (A.), Badrinath (B. R.). Recording distributed snapshots based on causal order of message delivery.Inf. Proc. Letters (1992),44, pp. 317–321.MATHCrossRefMathSciNetGoogle Scholar
  2. [2]
    Ahuja (M). Flush primitives for asynchronous distributed systems.Inf. Proc. Letters (1990),34, n° 2, pp. 5–12.MATHCrossRefMathSciNetGoogle Scholar
  3. [3]
    Ahuja (M.). Global snapshots for asynchronous distributed systems with non-FIFO channels.Tech. Rep. 92–268, Univ. of California, San Diego (Nov. 1992), 7 p.Google Scholar
  4. [4]
    Alagar (S.), Venkatesan (S.). An optimal algorithm for distributed snapshots with causal message ordering.Tech. Rep., Univ. of Texas, Dallas, Texas (1993), 7 p.Google Scholar
  5. [5]
    Babaoglu (O.), Marzullo (K.). Consistent global states of distributed systems: fondamental concepts and mechanisms. In S. J. Mullender, Editor, Distributed systems, Chap. 4,ACM Press (1993).Google Scholar
  6. [6]
    Birman (K.), Joseph (T. A.). Reliable communication in the presence of failures.ACM TDCS (Feb. 1987),5, n° 1, pp. 47–76.CrossRefGoogle Scholar
  7. [7]
    Chandy (K. M.), Lamport (L.). Distributed snapshots: determining global states of distributed systems.ACM TOCS (1985),3, n° 1, pp. 63–75.CrossRefGoogle Scholar
  8. [8]
    Chandy (K. M.), Misra (J.). Parallel program design: a foundation.Addison-Wesley (1988), 516 p.Google Scholar
  9. [9]
    Dukstra (E. W.). The distributed snapshot of K. M. Chandy and L. Lamport.Tech. Rep. EWD864a, Univ. of Texas, Austin, Tex. (1984).Google Scholar
  10. [10]
    Hélary (J.-M). Observing global states of asynchronous distributed applications.Proc. Workshop on Dist. Alg., Springer Verlag, LNCS 392 (1989), pp. 124–135.Google Scholar
  11. [11]
    Lai (T. H.), Yang (T. H.). On distributed snapshots.Inf. Proc. Letters (1987),25, pp. 153–158.MATHCrossRefMathSciNetGoogle Scholar
  12. [12]
    Lamport (L.). Time, clocks and the ordering of events in a distributed system.Comm. ACM (July 1978),21, n° 7, pp. 558–565.MATHCrossRefGoogle Scholar
  13. [13]
    Mattern (F.). Virtual time and global states of distributed systems.Proc. of Int. Workshop on Parallel and Dist. Systems, North Holland (1988), pp. 215–226.Google Scholar
  14. [14]
    Peterson (L. L.), Bucholz (N. C.), Schlichting (R.). Preserving and using context information in interprocess communication.ACM TOCS (1989),7, n° 3, pp. 213–246.CrossRefGoogle Scholar
  15. [15]
    Raynal (M.), Schiper (A.), Toueg (S.). The causal order abstraction and a simple way to implement it.Inf. Proc. Letters (1991),39, pp. 343–350.MATHCrossRefMathSciNetGoogle Scholar
  16. [16]
    Raynal (M.). Synchronisation et état global dans les systèmes répartis. Collection EDF,Eyrolles (1992), 230 p.Google Scholar
  17. [17]
    Sandoz (A.), Schiper (A.). A characterization of consistent distributed snapshot using causal order.Tech. Report RR 92–14, EPLF, Lausanne (Oct. 1992), 9 p.Google Scholar
  18. [18]
    Schiper (A.), Sandoz (A.), Eggli (J.). A new algorithm to implement causal ordering.Proc. Workshop on Dist. Alg., Springer Verlag, LNCS 392 (1989), pp. 219–232.Google Scholar
  19. [19]
    Taylor (K.). The role of inhibition in asynchronous consistent cut protocols.Proc. Workshop on Dist. Alg., Springer Verlag, LNCS392 (1989), pp. 280–291.Google Scholar

Copyright information

© Springer-Verlag 1994

Authors and Affiliations

  • Jean-Michel Hélary
    • 1
  • Achour Mostefaoui
    • 1
  • Michel Raynal
    • 1
  1. 1.IRISA Campus de BeaulieuRennes CedexFrance

Personalised recommendations