A distributed load-balancing policy for a multicomputer
Software—Practice & Experience
Epidemic algorithms for replicated database maintenance
PODC '87 Proceedings of the sixth annual ACM Symposium on Principles of distributed computing
Lightweight probabilistic broadcast
ACM Transactions on Computer Systems (TOCS)
Gossip-based aggregation in large dynamic networks
ACM Transactions on Computer Systems (TOCS)
Firefly-inspired Heartbeat Synchronization in Overlay Networks
SASO '07 Proceedings of the First International Conference on Self-Adaptive and Self-Organizing Systems
Gossiping in distributed systems
ACM SIGOPS Operating Systems Review - Gossip-based computer networking
T-Man: Gossip-based fast overlay topology construction
Computer Networks: The International Journal of Computer and Telecommunications Networking
| Hi-index | 0.00 |
Modern distributed systems may consist of hundreds of thousands of computers, ranging from high-end powerful machines to low-end resource-constrained wireless devices. We label them as extreme distributed systems, as they push scalability and complexity well beyond traditional scenarios. Most of these systems are still organized along traditional lines with hierarchical, centralized control planes. Things are changing though: more and more decentralized organizations are emerging, exemplified by P2P systems, ad-hoc networks, vehicular networks, etc. Decentralized organizations often combine local decision-making with dissemination of information in order to improve the decision-making process, exemplified by many gossip-based protocols. These protocols have been designed to solve problems as diverse as information dissemination, aggregation, topology maintenance, heartbeat synchronization, etc. Solutions to these problems share many common aspects, yet they have been published and developed in a confused and scattered way, leaving developers alone when integrating them into real applications. In this talk, we briefly introduce the gossip paradigm, showing how it can applied to solve several different problems, and we discuss what are the challenges that are left open for the researchers willing to provide a general framework for the design, implementation and deployment of gossip protocols.