Self-stabilizing resource discovery algorithm

Seda Davtyan, Kishori M. Konwar, Alexander A. Shvartsman

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Massive distributed cooperative computing in networks involves marshaling large collections of network nodes possessing the necessary computational resources. In order for the willing nodes to act in a concerted way they must first discover one another. This is the general setting of the Resource Discovery Problem (RDP). There are solutions for this problem that achieve impressive efficiency in the absence of failures, however, their correctness and performance cannot be guaranteed in the presence of failures. In practical environments it is important to have solutions that can cope with intermittent failures, and, in particular to design self-stabilizing algorithms for the problem. This paper presents a self-stabilizing algorithm that solves RDP in a deterministic synchronous setting. The approach is formulated in terms of evolving knowledge graphs, where vertices represent the participating network nodes, and edges represent one node's knowledge about another. Ideally, the diameter of such a graph is one, i.e., each node knows all others. The algorithm works in rounds as it evolves the knowledge graph by nodes sharing knowledge through gossip messages with the goal of reducing its diameter. We prove that the algorithm is self-stabilizing, that is, the algorithm is able to tolerate arbitrary perturbations in the nodes' local states and is guaranteed to solve the problem once such failures subside. We show that the algorithm has stabilization time of O(logD), and it takes at most 2logD + 10 complete round to stabilize, where D is the diameter of the initial knowledge graph. The corresponding message complexity is O(|V|2·logD), where V is the set of participating nodes.

Original languageEnglish (US)
Title of host publicationPrinciples of Distributed Systems - 17th International Conference, OPODIS 2013, Proceedings
Pages129-144
Number of pages16
DOIs
StatePublished - 2013
Externally publishedYes
Event17th International Conference on Principles of Distributed Systems, OPODIS 2013 - Rouen, France
Duration: Dec 16 2013Dec 18 2013

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume8304 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference17th International Conference on Principles of Distributed Systems, OPODIS 2013
Country/TerritoryFrance
CityRouen
Period12/16/1312/18/13

Keywords

  • Distributed Algorithm
  • Resource Discovery
  • Self-Stabilization

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Fingerprint

Dive into the research topics of 'Self-stabilizing resource discovery algorithm'. Together they form a unique fingerprint.

Cite this