An autonomic hierarchical reliable broadcast protocol for asynchronous distributed systems with failure detection

Élise Jeanneau,Luiz A Rodrigues,Luciana Arantes,Elias P Duarte Jr

doi:10.1186/s13173-017-0064-9

Abstract

Reliable broadcast is a fundamental building block in fault-tolerant distributed systems. It consists of a basic primitive that provides agreement among processes of the system on the delivery of each broadcast message, i.e., either none or all correct processes deliver the message, despite failures of processes. In this work, we propose a reliable broadcast solution on top of VCube, assuming that the system is asynchronous. VCube is an autonomic monitoring layer that organizes processes on a hypercube-like overlay which provides several logarithmic properties even in the presence of processes failures. We consider that processes can fail by crashing, do not recover, and faults are eventually detected by all correct processes. The protocol tolerates false suspicions by sending additional messages to suspected processes but logarithmic properties of the algorithm are still kept. Experimental results show the efficiency of the proposed solution compared to an one-to-all strategy.

Highlights

Broadcast is a basic component to implement numerous distributed applications and services such as notification, content delivery, replication, and group communication [1,2,3]
Reliable broadcast is defined in terms of two primitives: BROADCAST(m), which is defined by the broadcast algorithm and called by the application to disseminate m to all processes, and DELIVER(m), which is defined by the application and called by the broadcast algorithm when message m has been received
In a previous work [14], we proposed an autonomic reliable broadcast algorithm on top of the Hi-ADSD, a previous version of VCube

Summary

Introduction

Broadcast is a basic component to implement numerous distributed applications and services such as notification, content delivery, replication, and group communication [1,2,3]. A process in a distributed system uses broadcast to send a message to all other processes in the system. Reliable broadcast is defined in terms of two primitives: BROADCAST(m), which is defined by the broadcast algorithm and called by the application to disseminate m to all processes, and DELIVER(m), which is defined by the application and called by the broadcast algorithm when message m has been received. From an implementation point of view, the broadcast primitive sends point-to-point messages to each process of the system. Broadcast algorithms must ensure that, if a correct process broadcasts a message, it eventually delivers the message (validity property). Every correct process delivers a message at most once and only if that message was previously broadcast by some

Objectives

Methods

Discussion

Conclusion