Topology Aware P2P by Using DHT-based Flooding Approach

碩士 === 國立中興大學 === 資訊科學與工程學系 === 96 === P2P network is very popular in recent years, to resolve the traditional client server system in the provision of services brought about by the problem of heavy load. DHT-based systems les excel at supporting exact-match lookup; however, these systems do not sup...

Full description

Bibliographic Details
Main Authors: Hung-Yi Shiau, 蕭鴻益
Other Authors: 王丕中
Format: Others
Language:zh-TW
Online Access:http://ndltd.ncl.edu.tw/handle/93541167810838187150
id ndltd-TW-096NCHU5394086
record_format oai_dc
spelling ndltd-TW-096NCHU53940862016-05-09T04:13:51Z http://ndltd.ncl.edu.tw/handle/93541167810838187150 Topology Aware P2P by Using DHT-based Flooding Approach 一個基於使用DHT廣播搜尋的拓樸感知P2P網路演算法 Hung-Yi Shiau 蕭鴻益 碩士 國立中興大學 資訊科學與工程學系 96 P2P network is very popular in recent years, to resolve the traditional client server system in the provision of services brought about by the problem of heavy load. DHT-based systems les excel at supporting exact-match lookup; however, these systems do not support keyword-search. In contrast, unstructured P2P systems, like Guntella, support keyword-search, but the flooding-based algorithm does not scale. Moreover, P2P network is an overlay network which has been established over underlay network, so there is “Topology mismatch” problem. In this paper, we propose a topology-awareness P2P scheme to relieve the problem of topology mismatch. It can improve the performance of system broadcasting and decrease search latency. In this system, we categorize the topologically adjacent peers into one P2P group and use DHT-based flooding approach to broadcast query messages to the other group simultaneously to decrease the search latency. Finally, we evaluate the performance for our approach by Oversim simulator. The experimental results show that our approach can improve broadcasting performance and reduce search latency. 王丕中 學位論文 ; thesis 46 zh-TW
collection NDLTD
language zh-TW
format Others
sources NDLTD
description 碩士 === 國立中興大學 === 資訊科學與工程學系 === 96 === P2P network is very popular in recent years, to resolve the traditional client server system in the provision of services brought about by the problem of heavy load. DHT-based systems les excel at supporting exact-match lookup; however, these systems do not support keyword-search. In contrast, unstructured P2P systems, like Guntella, support keyword-search, but the flooding-based algorithm does not scale. Moreover, P2P network is an overlay network which has been established over underlay network, so there is “Topology mismatch” problem. In this paper, we propose a topology-awareness P2P scheme to relieve the problem of topology mismatch. It can improve the performance of system broadcasting and decrease search latency. In this system, we categorize the topologically adjacent peers into one P2P group and use DHT-based flooding approach to broadcast query messages to the other group simultaneously to decrease the search latency. Finally, we evaluate the performance for our approach by Oversim simulator. The experimental results show that our approach can improve broadcasting performance and reduce search latency.
author2 王丕中
author_facet 王丕中
Hung-Yi Shiau
蕭鴻益
author Hung-Yi Shiau
蕭鴻益
spellingShingle Hung-Yi Shiau
蕭鴻益
Topology Aware P2P by Using DHT-based Flooding Approach
author_sort Hung-Yi Shiau
title Topology Aware P2P by Using DHT-based Flooding Approach
title_short Topology Aware P2P by Using DHT-based Flooding Approach
title_full Topology Aware P2P by Using DHT-based Flooding Approach
title_fullStr Topology Aware P2P by Using DHT-based Flooding Approach
title_full_unstemmed Topology Aware P2P by Using DHT-based Flooding Approach
title_sort topology aware p2p by using dht-based flooding approach
url http://ndltd.ncl.edu.tw/handle/93541167810838187150
work_keys_str_mv AT hungyishiau topologyawarep2pbyusingdhtbasedfloodingapproach
AT xiāohóngyì topologyawarep2pbyusingdhtbasedfloodingapproach
AT hungyishiau yīgèjīyúshǐyòngdhtguǎngbōsōuxúndetàpǔgǎnzhīp2pwǎnglùyǎnsuànfǎ
AT xiāohóngyì yīgèjīyúshǐyòngdhtguǎngbōsōuxúndetàpǔgǎnzhīp2pwǎnglùyǎnsuànfǎ
_version_ 1718262889076228096