Efficient epidemic-style protocols for reliable and scalable multicast

Abstract

Epidemic-style (gossip-based) techniques have recently emerged as a scalable class of protocols for peer-to-peer reliable multicast dissemination in large process groups. These protocols provide probabilistic guarantees on reliability and scalability. However, popular implementations of epidemic-style dissemination are reputed to suffer from two major drawbacks: (a) (Network Overhead) when deployed on a WAN-wide or VPN-wide scale they generate a large number of packets that transit across the boundaries of multiple network domains (e.g., LANs, subnets, ASs), causing an overload on core network elements such as bridges, routers, and associated links; (b) (Lack of Adaptivity) they impose the same load on process group members and the network even under reduced failure rates (viz., packet losses, process failures). lit this paper we report on the (first) comprehensive set of solutions to these problems. The solution is comprised of two protocols: (1) a hierarchical gossiping protocol, and (2) an adaptive multicast dissemination framework that allows use of any gossiping primitive within it. These protocols work within a virtual peer-to-peer hierarchy called the Leaf Box hierarchy. Processes can be allocated in a topologically aware manner to the leaf boxes of this structure, so that (1) and (2) produce low traffic across domain boundaries in the network. In the interests of space, this paper focuses on a detailed discussion and evaluation (through simulations) of only the hierarchical gossiping protocol. We present an overview of the adaptive dissemination protocol and its properties.