Jaap-Henk Hoepman, Marina Papatrianta lou, and Philippas Tsigas. Self-stabilization of wait-free shared memory objects. Journal of Parallel and Distributed Computing, 62(5):766-791, 2002.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....the correct manner. The processing of communication variables, be they message Dolev s work was supported by BGU seed grant. Herman s work is sponsored by NSF award CAREER 97 9953. bu ers or communication registers, is atomic in these adaptations of [5] On the other hand, only few papers [17, 20, 14, 3]. address non atomic communication operations in the context of self stabilization. Lamport initially demonstrated that interprocess communication without explicity synchronization is possible [15] and formalizations of less than atomic communication were subsequently developed in [21, 16] The ....

....the unidirectional ring model associated with DijAlg. Our results are both positive (the constructions for regular and safe registers) and negative (the impossibility for 1W1R safe registers) Some similar experiences with the diculties of self stabilizing register constructions are reported in [14], however the problem adapting DijAlg has additional constraints and also advantages: the constraint of unidirectional communication rules out certain techniques, but the ring topology does provide sucient feedback (eventually information ows from p i back to p i 1 ) to make constructions ....

J.H. Hoepman, M. Papatrianta lou, and P. Tsigas, \Self-Stabilization in WaitFree Shared Memory Objects," Proceedings of the 9th International Workshop on Distributed Algorithms (WDAG '95), Lecture Notes in Computer Science Vol. 972, pages 273-287, Springer-Verlag, September 1995.

....Throughout this report we have been assuming systemic failures. Unifying self stabilization and fault tolerance is not a trivial problem. Can we have a system (protocol) that is self stabilizing, fault containing, and on top of that, fault tolerant as well Some results may already be available [19, 9, 16]; nevertheless this area is not yet fully explored. Our work can also be seen as a first step to understand the two popular models commonly used in distributed computing. It would be interesting to explore further fundamental differences between local shared memory and message passing. We can also ....

Jaap-Henk Hoepman, Marina Papatriantafilou, and Philippas Tsigas. Self-stabilization of wait-free shared memory objects. In Jean-Michel BIBLIOGRAPHY 80 H'elary and Michel Raynal, editors, 9th International Workshop on Distributed Algorithms, pages 273--287. Springer, 1995.

.... forgetful property [Sch93a] Variants of self stabilization have been proposed for control theoretic application [OWA91] and for situations where the initial state has a limited degree of arbitrariness [FMRT90] A variant may also be needed to address self stabilization of shared objects [LV92, HPT95] Hopfield nets and similar forms of neural networks may also have special definitions of the term [DWW86] 3 Advocacy, Surveys and Criticism The most frequently cited paper advocating self stabilization is Lamport s invited address [Lam83] Advantages of self stabilization are mentioned in a ....

....questions (uniformity, ring size) are mentioned in Sections 4 and 5. Self stabilizing algorithms and protocols can also be specified in models quite different from that of [Dij74] Petri nets [Tau89, Gho91b, CHR95, PM96] conventional shared memory systems [Lam86, LH91, ADHK97] wait free objects [HPT95] cellular automata [GKL78, Gac95] iteration systems [Gol82, AAEG93] rule based or constraint systems [BEG 90, CDK91, Che92, AGV94] neural networks [DWW86] genetic algorithms [DO93] and even digital [AG92] or continuous circuits [Hoo95] are represented in the bibliography. 10 Proof ....

JH Hoepman, M Papatriantafilou, and P Tsigas. Self-stabilization of wait-free shared memory objects. In WDAG95 Distributed Algorithms 9th International Workshop Proceedings, Springer-Verlag LNCS:972, pages 273--287, 1995.

No context found.

Jaap-Henk Hoepman, Marina Papatrianta lou, and Philippas Tsigas. Self-stabilization of wait-free shared memory objects. Journal of Parallel and Distributed Computing, 62(5):766-791, 2002.

No context found.

Jaap-Henk Hoepman, Marina Papatrianta lou, and Philippas Tsigas. Self-stabilization of wait-free shared memory objects. Journal of Parallel and Distributed Computing, 62(5):766-791, 2002.

....membership protocols) that can not only survive processor failures but also memory state corruptions. The latter type of errors have been extensively studied in the area of self stabilisation [Dij74] Algorithms that are both self stabilising and fault tolerant are much harder to design [AH93] HPT02] and this is even without any security requirements. ....

Hoepman, J.-H., Papatriantafilou, M., and Tsigas, P. Selfstabilization of wait-free shared memory objects. J. Parallel & Distr. Comput. 62, 5 (2002), 766--791.

....which the memory is faulty. In Research partially supported by the Dutch foundation for scienti c research (NWO) through NFI Proj. ALADDIN (contr. # NF 62 376) and a NUFFIC Fellowship, and by the EC ESPRIT II BRA Proj. ALCOM II (contr. # 7141) A preliminary version of this paper appeared as [HPT95] the rst model a distributed system must remain operational while a certain fraction of the processors is malfunctioning. When constructing shared memory objects like atomic registers, this issue is addressed by considering wait free constructions which guarantee that any operation executed by ....

....ones that satisfy this de nition. In this section we discuss possible variations on the de nition, and explain our particular choice of de nition. First, we initially thought that the corrupted actions could pose serious problems to the self stabilization of shared memory objects. Indeed, in [HPT95] we wrote Slow [corrupted] actions can carry the e ects of a transient error arbitrarily far into the future. Hence we can only say something meaningful about that part of a run after the time that all [corrupted] actions have nished, or the processors on which these [corrupted] actions run ....

[Article contains additional citation context not shown here]

Hoepman, J.-H., Papatriantafilou, M., and Tsigas, P. Self-stabilization of wait-free shared memory objects. In 9th Int. Workshop on Distributed Algorithms (Le Mont-Saint-Michel, France, 1995), J.-M. Helary and M. Raynal (Eds.), Lect. Not. Comp. Sci. 972, Springer, pp. 273-287.

....tolerance, distributed computing Research partially supported by the Dutch foundation for scientific research (NWO) through NFI Proj. ALADDIN (contr. # NF 62 376) and a NUFFIC Fellowship, and by the EC ESPRIT II BRA Proj. ALCOM II (contr. # 7141) A preliminary version of this paper appeared as [12]. Id: waitfree stab.tex,v 1.15 2002 01 30 17:45:14 hoepman Exp 1. INTRODUCTION In the past, research on fault tolerant distributed systems has focused either on system models in which processors fail, or on system models in which the memory is faulty. In the first model a distributed system ....

....ones that satisfy this definition. In this section we discuss possible variations on the definition, and explain our particular choice of definition. First, we initially thought that the corrupted actions could pose serious problems to the self stabilization of shared memory objects. Indeed, in [12] we wrote Slow [corrupted] actions can carry the e#ects of a transient error arbitrarily far into the future. Hence we can only say something meaningful about that part of a run after the time that all [corrupted] actions have finished, or the processors on which these [corrupted] actions run ....

[Article contains additional citation context not shown here]

J.-H. Hoepman, M. Papatriantafilou, and P. Tsigas, Self-stabilization of wait-free shared memory objects, in 9th Int. Workshop on Distributed Algorithms, J.-M. Helary and M. Raynal, eds., Lect. Not. Comp. Sci. 972, Le Mont-Saint-Michel, France, Sept. 12--15 1995, Springer, pp. 273--287.

....which the memory is faulty. In Research partially supported by the Dutch foundation for scienti c research (NWO) through NFI Proj. ALADDIN (contr. # NF 62 376) and a NUFFIC Fellowship, and by the EC ESPRIT II BRA Proj. ALCOM II (contr. # 7141) A preliminary version of this paper appeared as [HPT95] 1 the rst model a distributed system must remain operational while a certain fraction of the processors is malfunctioning. When constructing shared memory objects like atomic registers, this issue is addressed by considering wait free constructions which guarantee that any operation executed ....

....ones that satisfy this de nition. In this section we discuss possible variations on the de nition, and explain our particular choice of de nition. First, we initially thought that the corrupted actions could pose serious problems to the self stabilization of shared memory objects. Indeed, in [HPT95] we wrote Slow [corrupted] actions can carry the e ects of a transient error arbitrarily far into the future. Hence we can only say something meaningful about that part of a run after the time that all [corrupted] actions have nished, or the processors on which these [corrupted] actions run ....

[Article contains additional citation context not shown here]

Hoepman, J.-H., Papatriantafilou, M., and Tsigas, P. Self-stabilization of wait-free shared memory objects. In 9th Int. Workshop on Distributed Algorithms (Le Mont-Saint-Michel, France, 1995), J.-M. Helary and M. Raynal (Eds.), Lect. Not. Comp. Sci. 972, Springer, pp. 273-287.

....in which the memory is faulty. In Research partially supported by the Dutch foundation for scientific research (NWO) through NFI Proj. ALADDIN (contr. # NF 62 376) and a NUFFIC Fellowship, and by the EC ESPRIT II BRA Proj. ALCOM II (contr. # 7141) A preliminary version of this paper appeared as [HPT95] 1 the first model a distributed system must remain operational while a certain fraction of the processors is malfunctioning. When constructing shared memory objects like atomic registers, this issue is addressed by considering wait free constructions which guarantee that any operation ....

....ones that satisfy this definition. In this section we discuss possible variations on the definition, and explain our particular choice of definition. First, we initially thought that the corrupted actions could pose serious problems to the self stabilization of shared memory objects. Indeed, in [HPT95] we wrote S low [corrupted] actions can carry the effects of a transient error arbitrarily far into the future. Hence we can only say something meaningful about that part of a run after the time that all [corrupted] actions have finished, or the processors on which these [corrupted] actions ....

[Article contains additional citation context not shown here]

Hoepman, J.-H., Papatriantafilou, M., and Tsigas, P. Self-stabilization of wait-free shared memory objects. In 9th Int. Workshop on Distributed Algorithms (Le Mont-Saint-Michel, France, 1995), J.-M. H'elary and M. Raynal (Eds.), Lect. Not. Comp. Sci. 972, Springer, pp. 273--287.

No context found.

J-H Hoepman, M. Papatrianta lou, and P Tsigas. Self-stabilization of wait free shared memory objects. Journal of Parallel and Distributed Computing, 62(5):818-842, 2002.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC