A Hypercube-Based Scalable Peer-To-Peer Network
碩士 === 國立臺灣科技大學 === 電子工程系 === 93 === In order to provide for peer-to-peer networks efficient lookup and more load balancing, Distributed Hash Table (DHT) and various topologies of overlay network have become the fundamental parts to build peer-to-peer network systems. With its multi-dimensional, reg...
Main Authors: | , |
---|---|
Other Authors: | |
Format: | Others |
Language: | zh-TW |
Published: |
2005
|
Online Access: | http://ndltd.ncl.edu.tw/handle/76687493686287691414 |
id |
ndltd-TW-093NTUST428008 |
---|---|
record_format |
oai_dc |
spelling |
ndltd-TW-093NTUST4280082016-06-13T04:17:35Z http://ndltd.ncl.edu.tw/handle/76687493686287691414 A Hypercube-Based Scalable Peer-To-Peer Network 以超立方體為基礎的可擴充點對點網路 Kung-Yo Cheng 鄭功佑 碩士 國立臺灣科技大學 電子工程系 93 In order to provide for peer-to-peer networks efficient lookup and more load balancing, Distributed Hash Table (DHT) and various topologies of overlay network have become the fundamental parts to build peer-to-peer network systems. With its multi-dimensional, regular and symmetric features, hypercube architecture can provide messages routing with routing hops in average and efficient fault-tolerant. By employing the hypercube architecture, this paper proposes a hypercube-based peer-to-peer network (HBN), which possesses the characteristics of decentralizing, self-organizing and fault-tolerance, scalability, load balance and efficient lookup. According to the concept of uniform partitioning proposed in CAN (Content-Addressable Network), HBN can achieve more load balance. In addition, every node would backup its data to its predecessor node and successor node in a Hamiltonian cycle found in HBN according to Gray code order, resulting in enhancing the performance of recovering massive failed nodes. Hsing-Lung Cheng 陳省隆 2005 學位論文 ; thesis 43 zh-TW |
collection |
NDLTD |
language |
zh-TW |
format |
Others
|
sources |
NDLTD |
description |
碩士 === 國立臺灣科技大學 === 電子工程系 === 93 === In order to provide for peer-to-peer networks efficient lookup and more load balancing, Distributed Hash Table (DHT) and various topologies of overlay network have become the fundamental parts to build peer-to-peer network systems. With its multi-dimensional, regular and symmetric features, hypercube architecture can provide messages routing with routing hops in average and efficient fault-tolerant. By employing the hypercube architecture, this paper proposes a hypercube-based peer-to-peer network (HBN), which possesses the characteristics of decentralizing, self-organizing and fault-tolerance, scalability, load balance and efficient lookup. According to the concept of uniform partitioning proposed in CAN (Content-Addressable Network), HBN can achieve more load balance. In addition, every node would backup its data to its predecessor node and successor node in a Hamiltonian cycle found in HBN according to Gray code order, resulting in enhancing the performance of recovering massive failed nodes.
|
author2 |
Hsing-Lung Cheng |
author_facet |
Hsing-Lung Cheng Kung-Yo Cheng 鄭功佑 |
author |
Kung-Yo Cheng 鄭功佑 |
spellingShingle |
Kung-Yo Cheng 鄭功佑 A Hypercube-Based Scalable Peer-To-Peer Network |
author_sort |
Kung-Yo Cheng |
title |
A Hypercube-Based Scalable Peer-To-Peer Network |
title_short |
A Hypercube-Based Scalable Peer-To-Peer Network |
title_full |
A Hypercube-Based Scalable Peer-To-Peer Network |
title_fullStr |
A Hypercube-Based Scalable Peer-To-Peer Network |
title_full_unstemmed |
A Hypercube-Based Scalable Peer-To-Peer Network |
title_sort |
hypercube-based scalable peer-to-peer network |
publishDate |
2005 |
url |
http://ndltd.ncl.edu.tw/handle/76687493686287691414 |
work_keys_str_mv |
AT kungyocheng ahypercubebasedscalablepeertopeernetwork AT zhènggōngyòu ahypercubebasedscalablepeertopeernetwork AT kungyocheng yǐchāolìfāngtǐwèijīchǔdekěkuòchōngdiǎnduìdiǎnwǎnglù AT zhènggōngyòu yǐchāolìfāngtǐwèijīchǔdekěkuòchōngdiǎnduìdiǎnwǎnglù AT kungyocheng hypercubebasedscalablepeertopeernetwork AT zhènggōngyòu hypercubebasedscalablepeertopeernetwork |
_version_ |
1718304034921644032 |