2004 · 1.5K citations · 35 references
ComputingComputer ScienceComputer SystemsDistributed SystemsFault ToleranceFault-tolerant ComputingDistributed AlgorithmsDistributed ComputingFault-tolerant CommunicationInformation SpreadNetwork ReliabilityReliable Distributed SystemsAggregate FunctionsApproximation GuaranteesFault-tolerant Gossip ProtocolsGossip AlgorithmsInformation Dissemination
Recent advances in connectivity have shifted computing from centralized to highly distributed systems, yet large‑scale networks suffer frequent node and link failures; decentralized gossip protocols are emerging to maintain simplicity, scalability, and fault‑tolerant dissemination, with connections to random walk theory. This work investigates how gossip‑style protocols compute aggregates, presenting an analysis of simple gossip schemes for sums, averages, random samples, quantiles, and other functions that converge exponentially fast, and defining a precise diffusion speed metric. The authors analyze the diffusion speed of uniform gossip under node and link failures and compare it to flooding‑based mechanisms. They demonstrate that diffusion speed underpins the approximation guarantees for all considered aggregate problems.
Over the last decade, we have seen a revolution in connectivity between computers, and a resulting paradigm shift from centralized to highly distributed systems. With massive scale also comes massive instability, as node and link failures become the norm rather than the exception. For such highly volatile systems, decentralized gossip-based protocols are emerging as an approach to maintaining simplicity and scalability while achieving fault-tolerant information dissemination. In this paper, we study the problem of computing aggregates with gossip-style protocols. Our first contribution is an analysis of simple gossip-based protocols for the computation of sums, averages, random samples, quantiles, and other aggregate functions, and we show that our protocols converge exponentially fast to the true answer when using uniform gossip. Our second contribution is the definition of a precise notion of the speed with which a node's data diffuses through the network. We show that this diffusion speed is at the heart of the approximation guarantees for all of the above problems. We analyze the diffusion speed of uniform gossip in the presence of node and link failures, as well as for flooding-based mechanisms. The latter expose interesting connections to random walks on graphs.
35
Ion Stoica, Robert Morris, David R. Karger et al. · 2001 · 9.7K citations
Samuel Madden, Michael J. Franklin, Joseph M. Hellerstein et al. · ACM SIGOPS Operating Systems Review · 2002 · 2.7K citations