Computing along the routes via gossiping

Mehmet E. Yildiz, Anna Scaglione, Tuncer C. Aysal

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

3 Citations (Scopus)

Abstract

In this paper, we focus on a class of information propagation problems where a group of nodes (destinations) wants to gather a function of data that are stored by another disjoint group of nodes (sources). We assume that the function can be decomposed as a sum of functions of local variables. We study an algorithm which computes the function along the routes from sources to destinations iteratively, via gossiping among near neighbors. More specifically, in our protocol each node updates its own local value as a linear combination of its neighbors' values, repeatedly. Given the underlying network connectivity and the source-destination sets, we provide necessary and sufficient conditions for the feasibility of non-negative update weights. We further introduce topology based necessary conditions for the feasibility and discuss about known infeasible scenarios. By introducing the notion of source clusters, we also show that the complexity of the design problem is not directly related to the number of source nodes but the number of source clusters in the network.

Original languageEnglish (US)
Title of host publication2009 IEEE Information Theory Workshop, ITW 2009
Pages450-454
Number of pages5
DOIs
StatePublished - Dec 7 2009
Externally publishedYes
Event2009 IEEE Information Theory Workshop, ITW 2009 - Taormina, Sicily, Italy
Duration: Oct 11 2009Oct 16 2009

Other

Other2009 IEEE Information Theory Workshop, ITW 2009
CountryItaly
CityTaormina, Sicily
Period10/11/0910/16/09

Fingerprint

Topology

Keywords

  • Distributed computing
  • Gossiping protocols
  • Information flow
  • Network coding

ASJC Scopus subject areas

  • Computational Theory and Mathematics
  • Computer Networks and Communications
  • Information Systems
  • Electrical and Electronic Engineering

Cite this

Yildiz, M. E., Scaglione, A., & Aysal, T. C. (2009). Computing along the routes via gossiping. In 2009 IEEE Information Theory Workshop, ITW 2009 (pp. 450-454). [5351205] https://doi.org/10.1109/ITW.2009.5351205

Computing along the routes via gossiping. / Yildiz, Mehmet E.; Scaglione, Anna; Aysal, Tuncer C.

2009 IEEE Information Theory Workshop, ITW 2009. 2009. p. 450-454 5351205.

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

Yildiz, ME, Scaglione, A & Aysal, TC 2009, Computing along the routes via gossiping. in 2009 IEEE Information Theory Workshop, ITW 2009., 5351205, pp. 450-454, 2009 IEEE Information Theory Workshop, ITW 2009, Taormina, Sicily, Italy, 10/11/09. https://doi.org/10.1109/ITW.2009.5351205
Yildiz ME, Scaglione A, Aysal TC. Computing along the routes via gossiping. In 2009 IEEE Information Theory Workshop, ITW 2009. 2009. p. 450-454. 5351205 https://doi.org/10.1109/ITW.2009.5351205
Yildiz, Mehmet E. ; Scaglione, Anna ; Aysal, Tuncer C. / Computing along the routes via gossiping. 2009 IEEE Information Theory Workshop, ITW 2009. 2009. pp. 450-454
@inproceedings{093199d81f164efd862739fbc4d75644,
title = "Computing along the routes via gossiping",
abstract = "In this paper, we focus on a class of information propagation problems where a group of nodes (destinations) wants to gather a function of data that are stored by another disjoint group of nodes (sources). We assume that the function can be decomposed as a sum of functions of local variables. We study an algorithm which computes the function along the routes from sources to destinations iteratively, via gossiping among near neighbors. More specifically, in our protocol each node updates its own local value as a linear combination of its neighbors' values, repeatedly. Given the underlying network connectivity and the source-destination sets, we provide necessary and sufficient conditions for the feasibility of non-negative update weights. We further introduce topology based necessary conditions for the feasibility and discuss about known infeasible scenarios. By introducing the notion of source clusters, we also show that the complexity of the design problem is not directly related to the number of source nodes but the number of source clusters in the network.",
keywords = "Distributed computing, Gossiping protocols, Information flow, Network coding",
author = "Yildiz, {Mehmet E.} and Anna Scaglione and Aysal, {Tuncer C.}",
year = "2009",
month = "12",
day = "7",
doi = "10.1109/ITW.2009.5351205",
language = "English (US)",
isbn = "9781424449835",
pages = "450--454",
booktitle = "2009 IEEE Information Theory Workshop, ITW 2009",

}

TY - GEN

T1 - Computing along the routes via gossiping

AU - Yildiz, Mehmet E.

AU - Scaglione, Anna

AU - Aysal, Tuncer C.

PY - 2009/12/7

Y1 - 2009/12/7

N2 - In this paper, we focus on a class of information propagation problems where a group of nodes (destinations) wants to gather a function of data that are stored by another disjoint group of nodes (sources). We assume that the function can be decomposed as a sum of functions of local variables. We study an algorithm which computes the function along the routes from sources to destinations iteratively, via gossiping among near neighbors. More specifically, in our protocol each node updates its own local value as a linear combination of its neighbors' values, repeatedly. Given the underlying network connectivity and the source-destination sets, we provide necessary and sufficient conditions for the feasibility of non-negative update weights. We further introduce topology based necessary conditions for the feasibility and discuss about known infeasible scenarios. By introducing the notion of source clusters, we also show that the complexity of the design problem is not directly related to the number of source nodes but the number of source clusters in the network.

AB - In this paper, we focus on a class of information propagation problems where a group of nodes (destinations) wants to gather a function of data that are stored by another disjoint group of nodes (sources). We assume that the function can be decomposed as a sum of functions of local variables. We study an algorithm which computes the function along the routes from sources to destinations iteratively, via gossiping among near neighbors. More specifically, in our protocol each node updates its own local value as a linear combination of its neighbors' values, repeatedly. Given the underlying network connectivity and the source-destination sets, we provide necessary and sufficient conditions for the feasibility of non-negative update weights. We further introduce topology based necessary conditions for the feasibility and discuss about known infeasible scenarios. By introducing the notion of source clusters, we also show that the complexity of the design problem is not directly related to the number of source nodes but the number of source clusters in the network.

KW - Distributed computing

KW - Gossiping protocols

KW - Information flow

KW - Network coding

UR - http://www.scopus.com/inward/record.url?scp=76249109669&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=76249109669&partnerID=8YFLogxK

U2 - 10.1109/ITW.2009.5351205

DO - 10.1109/ITW.2009.5351205

M3 - Conference contribution

SN - 9781424449835

SP - 450

EP - 454

BT - 2009 IEEE Information Theory Workshop, ITW 2009

ER -