An Efficient Multicast Routing Algorithm in BCube Networks

碩士 === 國立清華大學 === 資訊工程學系 === 102 === In recent years, cloud computing becomes more and more popular. BCube is one of the novel architecture proposed for data centers. However, finding a multicast tree in BCube is a Steiner tree problem which is an NP-hard problem. We propose a heuristic algorithm, R...

Full description

Bibliographic Details
Main Authors: Lo, Tzu-Ting, 羅子庭
Other Authors: Sheu, Jang-Ping
Format: Others
Language:en_US
Published: 2014
Online Access:http://ndltd.ncl.edu.tw/handle/75897682384971448107
id ndltd-TW-102NTHU5392130
record_format oai_dc
spelling ndltd-TW-102NTHU53921302016-03-09T04:34:23Z http://ndltd.ncl.edu.tw/handle/75897682384971448107 An Efficient Multicast Routing Algorithm in BCube Networks 在BCube拓樸上的一個有效率多點傳播路由演算法 Lo, Tzu-Ting 羅子庭 碩士 國立清華大學 資訊工程學系 102 In recent years, cloud computing becomes more and more popular. BCube is one of the novel architecture proposed for data centers. However, finding a multicast tree in BCube is a Steiner tree problem which is an NP-hard problem. We propose a heuristic algorithm, Resource-Saving Multicast Tree (RSMT) algorithm. The goal of RSMT is to find a multicast tree with high path reusability for a multicast group. We design a greedy method that picks a local optimal receiver and union it to multicast tree at a time. RSMT also can handle the dynamic joining and leaving of receivers by reconstructing the multicast tree when it is no longer efficient. The experiment result shows that we can find a multicast tree with lower communication cost and higher throughput compared to existed solutions. Furthermore, we observe that we can sacrifice some delay-time of some receivers to reduce the total communication cost of a multicast tree for small BCube with small multicast group size. Sheu, Jang-Ping 許健平 2014 學位論文 ; thesis 31 en_US
collection NDLTD
language en_US
format Others
sources NDLTD
description 碩士 === 國立清華大學 === 資訊工程學系 === 102 === In recent years, cloud computing becomes more and more popular. BCube is one of the novel architecture proposed for data centers. However, finding a multicast tree in BCube is a Steiner tree problem which is an NP-hard problem. We propose a heuristic algorithm, Resource-Saving Multicast Tree (RSMT) algorithm. The goal of RSMT is to find a multicast tree with high path reusability for a multicast group. We design a greedy method that picks a local optimal receiver and union it to multicast tree at a time. RSMT also can handle the dynamic joining and leaving of receivers by reconstructing the multicast tree when it is no longer efficient. The experiment result shows that we can find a multicast tree with lower communication cost and higher throughput compared to existed solutions. Furthermore, we observe that we can sacrifice some delay-time of some receivers to reduce the total communication cost of a multicast tree for small BCube with small multicast group size.
author2 Sheu, Jang-Ping
author_facet Sheu, Jang-Ping
Lo, Tzu-Ting
羅子庭
author Lo, Tzu-Ting
羅子庭
spellingShingle Lo, Tzu-Ting
羅子庭
An Efficient Multicast Routing Algorithm in BCube Networks
author_sort Lo, Tzu-Ting
title An Efficient Multicast Routing Algorithm in BCube Networks
title_short An Efficient Multicast Routing Algorithm in BCube Networks
title_full An Efficient Multicast Routing Algorithm in BCube Networks
title_fullStr An Efficient Multicast Routing Algorithm in BCube Networks
title_full_unstemmed An Efficient Multicast Routing Algorithm in BCube Networks
title_sort efficient multicast routing algorithm in bcube networks
publishDate 2014
url http://ndltd.ncl.edu.tw/handle/75897682384971448107
work_keys_str_mv AT lotzuting anefficientmulticastroutingalgorithminbcubenetworks
AT luózitíng anefficientmulticastroutingalgorithminbcubenetworks
AT lotzuting zàibcubetàpǔshàngdeyīgèyǒuxiàolǜduōdiǎnchuánbōlùyóuyǎnsuànfǎ
AT luózitíng zàibcubetàpǔshàngdeyīgèyǒuxiàolǜduōdiǎnchuánbōlùyóuyǎnsuànfǎ
AT lotzuting efficientmulticastroutingalgorithminbcubenetworks
AT luózitíng efficientmulticastroutingalgorithminbcubenetworks
_version_ 1718202159888072704