Peer-to-Peer Membership Management for Gossip-Based Protocols

@article{Ganesh2003PeertoPeerMM,
title={Peer-to-Peer Membership Management for Gossip-Based Protocols},
author={Ayalvadi J. Ganesh and Anne-Marie Kermarrec and Laurent Massouli{\'e}},
journal={IEEE Trans. Computers},
year={2003},
volume={52},
pages={139-149}
}
• Published 1 February 2003
• Computer Science
• IEEE Trans. Computers
Gossip-based protocols for group communication have attractive scalability and reliability properties. The probabilistic gossip schemes studied so far typically assume that each group member has full knowledge of the global membership and chooses gossip targets uniformly at random. The requirement of global knowledge impairs their applicability to very large-scale groups. In this paper, we present SCAMP (Scalable Membership protocol), a novel peer-to-peer membership protocol which operates in a…
538 Citations

Figures and Tables from this paper

Correctness of a gossip based membership protocol
• Computer Science
PODC '05
• 2005
A new scalable gossip-based algorithm for local view maintenance is given, together with a proof that the expected time until a network partition is at least exponential in the square of the view size, and probabilistic bounds on the in-degree are developed.
Gossip-based peer sampling
• Computer Science
TOCS
• 2007
This paper presents a generic framework to implement a peer-sampling service in a decentralized manner by constructing and maintaining dynamic unstructured overlays through gossiping membership information itself, which generalizes existing approaches and makes it easy to discover new ones.
On the Correctness of Gossip-based Membership Protocols
A new scalable gossip-based algorithm for local view maintenance is given, together with a proof that the expected time until a network partition is at least exponential in the view size and the size of the departing set, and probabilistic bounds on the in-degree of individual nodes are developed.
DHT-Aid, Gossip-Based Heterogeneous Peer-to-Peer Membership Management
• Computer Science
2008 5th IEEE Consumer Communications and Networking Conference
• 2008
A DHT-aid, gossip-based heterogeneous peer-to-peer membership protocol, called DIGOM, which achieves a good quality of load balance and requires no synchronization and can satisfy the requirements for reliable dissemination.
Gossip-based Unstructured Overlay Networks: An Experimental Evaluation
• Computer Science
• 2003
This paper generalizes existing gossip-based overlays by introducing a general scheme in which existing overlays as well as novel protocols can be implemented, and shows that all of them lead to different communication topologies none of which is uniformly random.
The Peer Sampling Service: Experimental Evaluation of Unstructured Gossip-Based Implementations
• Computer Science
Middleware
• 2004
This paper presents a generic framework to implement reliable and efficient peer sampling services, which generalizes existing approaches and makes it easy to introduce new ones, and shows that all of them lead to differentpeer sampling services none of which is uniformly random.
Novel Gossip-Based Multicast Protocol with No Global View Information
• Computer Science
2008 Second International Conference on Future Generation Communication and Networking
• 2008
An efficient gossip-based multicast protocol to guarantee causally-ordered delivery semantics based on the local view of every individual member consisting of a subset of members continuously changing, whose size never excesses a determined threshold, without virtual synchrony and K-committee.
HyParView: A Membership Protocol for Reliable Gossip-Based Broadcast
• Computer Science
37th Annual IEEE/IFIP International Conference on Dependable Systems and Networks (DSN'07)
• 2007
This paper presents HyParView, a new membership protocol to support gossip-based broadcast that ensures high levels of reliability even in the presence of high rates of node failure.
Simulations results show that, in comparison with other existing protocols, HyParView-based gossip protocols not only provide better reliability but also support higher percentages of node failures, and are able to recover faster from these failures.
Probabilistic Reliable Dissemination in Large-Scale Systems
• Computer Science
IEEE Trans. Parallel Distributed Syst.
• 2003
This paper provides a theoretical analysis of gossip-based protocols which relates their reliability to key system parameters (the system size, failure rates, and number of gossip targets) and shows how reliability can be maintained while alleviating drawback.

References

SHOWING 1-10 OF 36 REFERENCES
SCAMP: Peer-to-Peer Lightweight Membership Service for Large-Scale Group Communication
• Computer Science
Networked Group Communication
• 2001
This paper presents a novel peer-to-peer membership service which operates in a completely decentralized manner in that nobody has global knowledge of membership, and demonstrates that membership information is replicated robustly enough to support gossip with high reliability.
Probabilistic Reliable Dissemination in Large-Scale Systems
• Computer Science
IEEE Trans. Parallel Distributed Syst.
• 2003
This paper provides a theoretical analysis of gossip-based protocols which relates their reliability to key system parameters (the system size, failure rates, and number of gossip targets) and shows how reliability can be maintained while alleviating drawback.
Directional Gossip: Gossip in a Wide Area Network
• Computer Science
EDCC
• 1999
A new gossip protocol is presented, called {\em directional gossip}, that uses flooding when necessary to attain good reliability and that uses gossip when flooding (and hence its inherent high overhead) is not needed.
A gossip-based reliable multicast for large-scale high-throughput applications
• Computer Science
Proceeding International Conference on Dependable Systems and Networks. DSN 2000
• 2000
This paper presents a reliable probabilistic multicast, rpbcast, that is a hybrid of the centralized and gossip-based approaches and shows that negative gossip allows pull based recovery and converges faster than push based recovery.
Scribe: a large-scale and decentralized application-level multicast infrastructure
• Computer Science
IEEE J. Sel. Areas Commun.
• 2002
Simulation results, based on a realistic network topology model, show that Scribe scales across a wide range of groups and group sizes, and balances the load on the nodes while achieving acceptable delay and link stress when compared with Internet protocol multicast.
Scalable group membership services for novel applications
• Computer Science
Networks in Distributed Computing
• 1997
A new architecture for a scalable group membership service for wide area environments that supports the coexistence of full-edged clients along with thin clients and the novelty of the design is in its client-server approach, which allows lightweight clients to beneet from advanced membership services.
A client-server oriented algorithm for virtually synchronous group membership in WANs
• Computer Science
Proceedings 20th IEEE International Conference on Distributed Computing Systems
• 2000
This work describes a novel scalable group membership service designed explicitly for wide area networks that attains, on average, a low message overhead by agreeing on membership within a single message round.
Gossip versus Deterministically Constrained Flooding on Small Networks
• Computer Science
DISC
• 2000
This paper presents a protocol that superficially resembles rumor mongering but is deterministic, and shows that this new protocol has most of the same attractions as rumormongering.
Application-Level Multicast Using Content-Addressable Networks
• Computer Science
Networked Group Communication
• 2001
This paper proposes an application-level multicast scheme capable of scaling to large group sizes without restricting the service model to a single source, and believes it offers the dual advantages of simplicity and scalability.
Bimodal multicast
• Computer Science
TOCS
• 1999
This article introduces the protocol, provides a theoretical analysis of its behavior, review experimental results, and discusses some candidate applications, confirming that bimodal multicast is reliable, scalable, and that the protocol provides remarkably stable delivery throughput.