An Autonomic Implementation of Reliable Broadcast Based on Dynamic Spanning Trees

Abstract

Reliable broadcast is a basic building block of dependable distributed systems that allows the dissemination of messages to all processes guaranteeing that either none or all correct processes deliver the message, despite the occurrence of failures. In this work we introduce an autonomic strategy to implement reliable broadcast. Processes self-organize themselves on an overlay based on a hypercube-like topology. The overlay is maintained by the execution of an underlying monitoring algorithm that guarantees several logarithmic properties even in the presence of processes failures. The reliable broadcast strategy employs spanning trees that are dynamically built embedded in the overlay. Nodes fail by crashing, crashes are permanent, and the broadcast strategy works correctly for an arbitrary number of process crashes. After a process crashes, the reconstruction of the spanning trees is transparent, not inducing, therefore, performance degradation. Besides the formal specification, we also present experimental results obtained with simulation.

Bundling Messages to Reduce the Cost of Tree-Based Broadcast Algorithms

“…Using the above functions, the following algorithm, proposed in [8], creates a spanning tree, rooted at i, propagating m over all non faulty nodes of the system:…”

“…One way to mitigate these problems is to use multiple adaptive trees. In [7], [8], [9] we proposed broadcast algorithms based on the VCube [10]. VCube is hypercubelike virtual topology that reconfigures itself and maintains multiple logarithmic properties even after processes crash.…”

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

“…In [14], we presented a reliable broadcast solution based on dynamic spanning trees on top of the Hi-ADSD, a previous version of the VCube. Multiple trees are dynamically built that include all correct nodes, where each tree root corresponds to the node that called a broadcast primitive.…”

“…In a previous work [14], we proposed an autonomic reliable broadcast algorithm on top of the Hi-ADSD, a previous version of VCube. The algorithm guarantees several logarithmic properties, even when nodes fail and allows transparent and efficient spanning tree reconstructions.…”

VCube-PS: A causal broadcast topic-based publish/subscribe system

“…On the other hand, p 7 verifies that in cluster c 7,2 = (5, 4), p 5 is a subscriber of t 2 (Children(7, t 2 , 2) = {5}), and therefore sends m 2 to p 5 which on its turn does not send it to p 4 , because even if p 4 is the first and only node in c 5,1 , it is not a subscriber of t 2 . For more details about how to build spanning trees over VCube, see [18].…”