Virtually-synchronous communication based on a weak failure suspector

Schiper, André; Ricciardi, Aleta

doi:10.1109/ftcs.1993.627356

Cited by 33 publications

(10 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Reachability information is typically derived from a system service called the failure 8uspector [6,15]. In an asynchronous system, conclusions regarding reachability derived by individual processes can never be totally accurate and may be mutually inconsistent.…”

Section: Large-scale Distributed Systemsmentioning

confidence: 99%

See 1 more Smart Citation

On group communication in large-scale distributed systems

Babaoğlu

Schiper

1995

SIGOPS Oper. Syst. Rev.

View full text Add to dashboard Cite

An increasing number of applications with reliability requirements are being deployed in distributed systems that span large geographic distances or manage large numbers of objects. We consider the process group mechanism as an appropriate application structuring paradigm in such large-scale distributed systems. We give a formal characterization for the attribute "large scale" as applied to distributed systems and examine the technical problems that need to be solved in making group technology scalable. Our design advocates multiple roles for group membership over a minimal set of abstractions and primitives. The design is currently being implemented on top of "off-the-shelf" technologies for both communication and computation.

show abstract

Section: Large-scale Distributed Systemsmentioning

confidence: 99%

“…We believe that the ability to manage shared state in an asynchronous large-scale distributed system requires a minimum set of fault-tolerant abstractions similar to those defined by view-s~lnchronous commgnication (VSC) [15,1]. 1 The basic primitive of VSC is the reliable multicast of a message to a group of processes.…”

Section: Core Membersmentioning

confidence: 99%

On group communication in large-scale distributed systems

Babaoğlu

Schiper

1995

SIGOPS Oper. Syst. Rev.

View full text Add to dashboard Cite

show abstract

“…At the lowest layer, we rely on a view-synchronoua communication (VSC) service [15,1,5] providing a minimal set of primitives for reliable multicasts and membership changes. The core members rely on VSC semantics for internal communication and the handling of request/reply interactions with clients.…”

Section: Introductionmentioning

confidence: 99%

On group communication in large-scale distributed systems

Babaoğlu

Schiper

1994

Proceedings of the 6th Workshop on ACM SIGOPS European Workshop: Matching Operating Systems to Application Needs

View full text Add to dashboard Cite

show abstract

“…Local Monotonicity has two important consequences: it guarantees that the same view is not delivered more then once to the same end-point and that if two views are delivered to two end-points they are delivered in the same order. These properties are ful lled by virtually all group membership services (e.g., 12,16,7,19,10,27,34,5]). …”

Section: The Membership Service Speci Cationmentioning

confidence: 99%

“…Many variants of virtual synchrony s e m a n tics have been suggested 32,19,39,12,34,18]. All of these variants specify that every message is delivered in the same view by all processes that deliver it.…”

Section: Introductionmentioning

confidence: 99%

A client-server approach to virtually synchronous group multicast: specifications and algorithms

Keidar¹,

Khazan

Proceedings 20th IEEE International Conference on Distributed Computing Systems

View full text Add to dashboard Cite

This paper presents a formal design for a novel group multicast service that provides virtually synchronous semantics in asynchronous fault-prone environments. The design employs a clientserver architecture in which group membership is maintained not by e v ery process but only by dedicated membership servers, while virtually synchronous group multicast is implemented by service end-points running at the clients. This architecture allows the service to be scalable in the topology it spans, in the number of groups, and in the number of clients. Our design allows the virtual synchrony algorithm to run in a single message exchange round, in parallel with the membership algorithm: it does not require pre-agreement upon a common identi er by the membership algorithm.Speci cally, the paper de nes service semantics for the client-server interface, that is, for the group membership service. The paper then speci es virtually synchronous semantics for the new group multicast service, as a collection of safety a n d l i v eness properties. These properties have been previously suggested and have been shown to be useful for distributed applications. The paper then presents new algorithms that use the de ned group membership service to implement the speci ed properties. The speci cations and algorithms are presented incrementally, using a novel inheritance-based formal construct 26]. The algorithm that provides the complete virtually synchronous semantics executes in a single message round, and is therefore more e cient than previously suggested algorithms providing such semantics. The algorithm has been implemented in C++. All the speci cations and algorithms are presented using the I/O automaton formalism. Furthermore, the paper includes formal proofs showing that the algorithms meet their speci cations. Safety properties are proven using invariant assertions and simulations. Liveness is proven using invariant assertions and careful operational arguments.

show abstract

Virtually-synchronous communication based on a weak failure suspector

Cited by 33 publications

References 26 publications

On group communication in large-scale distributed systems

On group communication in large-scale distributed systems

On group communication in large-scale distributed systems

A client-server approach to virtually synchronous group multicast: specifications and algorithms

Contact Info

Product

Resources

About