Optimizing peer relationships in a super-peer network

Paweł Garbacki, Dick H.J. Epema, Maarten van Steen

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

24 Citations (Scopus)
6 Downloads (Pure)

Abstract

Super-peer architectures exploit the heterogeneity of nodes in a P2P network by assigning additional responsibilities to higher-capacity nodes. In the design of a super-peer network for file sharing, several issues have to be addressed: how client peers are related to super-peers, how super-peers locate files, how the load is balanced among the super-peers, and how the system deals with node failures. In this paper we introduce a self-organizing super-peer network architecture (SOSPNET) that solves these issues in a fully decentralized manner. SOSPNET maintains a super-peer network topology that reflects the semantic similarity of peers sharing content interests. Super-peers maintain semantic caches of pointers to files which are requested by peers with similar interests. Client peers, on the other hand, dynamically select super-peers offering the best search performance. We show how this simple approach can be employed not only to optimize searching, but also to solve generally difficult problems encountered in P2P architectures such as load balancing and fault tolerance. We evaluate SOSPNET using a model of the semantic structure derived from the 8-month traces of two large file-sharing communities. The obtained results indicate that SOSPNET achieves close-to-optimal file search performance, quickly adjusts to changes in the environment (node joins and leaves), survives even catastrophic node failures, and efficiently distributes the system load taking into account peer capacities.

Original languageEnglish
Title of host publication27th International Conference on Distributed Computing Systems, ICDCS'07
Subtitle of host publicationToronto ON, Canada, June 25-27, 2007
PublisherIEEE
Number of pages10
ISBN (Print)9780769528373
DOIs
Publication statusPublished - 8 Oct 2007
Externally publishedYes
Event27th International Conference on Distributed Computing Systems, ICDCS 2007 - Toronto, Canada
Duration: 25 Jun 200727 Jun 2007
Conference number: 27

Publication series

NameInternational Conference on Distributed Computing Systems (ICDCS)
PublisherIEEE
Volume2007
ISSN (Print)1063-6927

Conference

Conference27th International Conference on Distributed Computing Systems, ICDCS 2007
Abbreviated titleICDCS
CountryCanada
CityToronto
Period25/06/0727/06/07

Fingerprint

Network architecture
Semantics
Fault tolerance
Resource allocation
Topology

Keywords

  • Peer to peer computing
  • Network topology
  • Design optimization
  • Protocols
  • Load management
  • Fault tolerance
  • Network servers
  • Routing

Cite this

Garbacki, P., Epema, D. H. J., & van Steen, M. (2007). Optimizing peer relationships in a super-peer network. In 27th International Conference on Distributed Computing Systems, ICDCS'07: Toronto ON, Canada, June 25-27, 2007 [4268185] (International Conference on Distributed Computing Systems (ICDCS); Vol. 2007). IEEE. https://doi.org/10.1109/ICDCS.2007.126
Garbacki, Paweł ; Epema, Dick H.J. ; van Steen, Maarten. / Optimizing peer relationships in a super-peer network. 27th International Conference on Distributed Computing Systems, ICDCS'07: Toronto ON, Canada, June 25-27, 2007. IEEE, 2007. (International Conference on Distributed Computing Systems (ICDCS)).
@inproceedings{11d4e2e675614c9089f78eea9fdcdf22,
title = "Optimizing peer relationships in a super-peer network",
abstract = "Super-peer architectures exploit the heterogeneity of nodes in a P2P network by assigning additional responsibilities to higher-capacity nodes. In the design of a super-peer network for file sharing, several issues have to be addressed: how client peers are related to super-peers, how super-peers locate files, how the load is balanced among the super-peers, and how the system deals with node failures. In this paper we introduce a self-organizing super-peer network architecture (SOSPNET) that solves these issues in a fully decentralized manner. SOSPNET maintains a super-peer network topology that reflects the semantic similarity of peers sharing content interests. Super-peers maintain semantic caches of pointers to files which are requested by peers with similar interests. Client peers, on the other hand, dynamically select super-peers offering the best search performance. We show how this simple approach can be employed not only to optimize searching, but also to solve generally difficult problems encountered in P2P architectures such as load balancing and fault tolerance. We evaluate SOSPNET using a model of the semantic structure derived from the 8-month traces of two large file-sharing communities. The obtained results indicate that SOSPNET achieves close-to-optimal file search performance, quickly adjusts to changes in the environment (node joins and leaves), survives even catastrophic node failures, and efficiently distributes the system load taking into account peer capacities.",
keywords = "Peer to peer computing, Network topology, Design optimization, Protocols, Load management, Fault tolerance, Network servers, Routing",
author = "Paweł Garbacki and Epema, {Dick H.J.} and {van Steen}, Maarten",
year = "2007",
month = "10",
day = "8",
doi = "10.1109/ICDCS.2007.126",
language = "English",
isbn = "9780769528373",
series = "International Conference on Distributed Computing Systems (ICDCS)",
publisher = "IEEE",
booktitle = "27th International Conference on Distributed Computing Systems, ICDCS'07",
address = "United States",

}

Garbacki, P, Epema, DHJ & van Steen, M 2007, Optimizing peer relationships in a super-peer network. in 27th International Conference on Distributed Computing Systems, ICDCS'07: Toronto ON, Canada, June 25-27, 2007., 4268185, International Conference on Distributed Computing Systems (ICDCS), vol. 2007, IEEE, 27th International Conference on Distributed Computing Systems, ICDCS 2007, Toronto, Canada, 25/06/07. https://doi.org/10.1109/ICDCS.2007.126

Optimizing peer relationships in a super-peer network. / Garbacki, Paweł; Epema, Dick H.J.; van Steen, Maarten.

27th International Conference on Distributed Computing Systems, ICDCS'07: Toronto ON, Canada, June 25-27, 2007. IEEE, 2007. 4268185 (International Conference on Distributed Computing Systems (ICDCS); Vol. 2007).

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

TY - GEN

T1 - Optimizing peer relationships in a super-peer network

AU - Garbacki, Paweł

AU - Epema, Dick H.J.

AU - van Steen, Maarten

PY - 2007/10/8

Y1 - 2007/10/8

N2 - Super-peer architectures exploit the heterogeneity of nodes in a P2P network by assigning additional responsibilities to higher-capacity nodes. In the design of a super-peer network for file sharing, several issues have to be addressed: how client peers are related to super-peers, how super-peers locate files, how the load is balanced among the super-peers, and how the system deals with node failures. In this paper we introduce a self-organizing super-peer network architecture (SOSPNET) that solves these issues in a fully decentralized manner. SOSPNET maintains a super-peer network topology that reflects the semantic similarity of peers sharing content interests. Super-peers maintain semantic caches of pointers to files which are requested by peers with similar interests. Client peers, on the other hand, dynamically select super-peers offering the best search performance. We show how this simple approach can be employed not only to optimize searching, but also to solve generally difficult problems encountered in P2P architectures such as load balancing and fault tolerance. We evaluate SOSPNET using a model of the semantic structure derived from the 8-month traces of two large file-sharing communities. The obtained results indicate that SOSPNET achieves close-to-optimal file search performance, quickly adjusts to changes in the environment (node joins and leaves), survives even catastrophic node failures, and efficiently distributes the system load taking into account peer capacities.

AB - Super-peer architectures exploit the heterogeneity of nodes in a P2P network by assigning additional responsibilities to higher-capacity nodes. In the design of a super-peer network for file sharing, several issues have to be addressed: how client peers are related to super-peers, how super-peers locate files, how the load is balanced among the super-peers, and how the system deals with node failures. In this paper we introduce a self-organizing super-peer network architecture (SOSPNET) that solves these issues in a fully decentralized manner. SOSPNET maintains a super-peer network topology that reflects the semantic similarity of peers sharing content interests. Super-peers maintain semantic caches of pointers to files which are requested by peers with similar interests. Client peers, on the other hand, dynamically select super-peers offering the best search performance. We show how this simple approach can be employed not only to optimize searching, but also to solve generally difficult problems encountered in P2P architectures such as load balancing and fault tolerance. We evaluate SOSPNET using a model of the semantic structure derived from the 8-month traces of two large file-sharing communities. The obtained results indicate that SOSPNET achieves close-to-optimal file search performance, quickly adjusts to changes in the environment (node joins and leaves), survives even catastrophic node failures, and efficiently distributes the system load taking into account peer capacities.

KW - Peer to peer computing

KW - Network topology

KW - Design optimization

KW - Protocols

KW - Load management

KW - Fault tolerance

KW - Network servers

KW - Routing

UR - http://www.scopus.com/inward/record.url?scp=34848849815&partnerID=8YFLogxK

U2 - 10.1109/ICDCS.2007.126

DO - 10.1109/ICDCS.2007.126

M3 - Conference contribution

SN - 9780769528373

T3 - International Conference on Distributed Computing Systems (ICDCS)

BT - 27th International Conference on Distributed Computing Systems, ICDCS'07

PB - IEEE

ER -

Garbacki P, Epema DHJ, van Steen M. Optimizing peer relationships in a super-peer network. In 27th International Conference on Distributed Computing Systems, ICDCS'07: Toronto ON, Canada, June 25-27, 2007. IEEE. 2007. 4268185. (International Conference on Distributed Computing Systems (ICDCS)). https://doi.org/10.1109/ICDCS.2007.126