|
English
|
正體中文
|
简体中文
|
Items with full text/Total items : 21921/27947 (78%)
Visitors : 4216437
Online Users : 345
|
|
|
Loading...
|
Please use this identifier to cite or link to this item:
http://140.128.103.80:8080/handle/310901/24599
|
Title: | 建構一個植基於多使用者星狀拓樸資料網格的群播協同檔案傳送機制 |
Other Titles: | The Construction of a Co-Allocation Scheme Based on a Star-Topology Data Grid with Multiple Users |
Authors: | 呂芳懌 ?LEU, FANG-YIE |
Contributors: | 東海大學資訊工程學系 行政院國家科學委員會 |
Date: | 2012 |
Issue Date: | 2014-03-07T07:55:07Z (UTC)
|
Abstract: | 資料網格(data grid)為一網格系統(Grid system)之擴充版本,它可以整合分散在不同地理位置之計算資源與儲存資源,以茲服務使用者。同時,隨著大規模且複雜的研究持續地進行,資料網格中的網路拓樸與工作站也因而持續不斷地擴張,使得分析一個新的資源仲介代理人(Resource Broker)排程程式之績效或分析平行地由各地工作站傳送某一檔案之片段,給其使用者之網路行為異常困難,即,協同傳送機制(co-allocation scheme)。為此,我們提出一個化簡網路拓樸之方法,稱之為Tree-to-Star Reduction method (簡稱為TSR),目標是將一個複雜的樹狀網路拓樸化簡成一個等效的星狀拓樸,以降低研究人員在分析協同傳送機制之網路行為時的負擔,同時TSR也可以用來化簡其它平行分散式系統之樹狀網路拓樸,例如:網格系統、雲端計算與P2P系統,以利於分析檔案傳遞時間(job turnaround time)等,而不失其準確性。 A data grid, the extension of a Grid system, often integrates geographically dispersed computational and storage resources. Users are able to execute large-scale applications and store their experimental data and/or produced data in various areas in the data girds, such as physics, astronomy, and seismology. However, it is difficult for researchers to evaluate the performance of proposed algorithm since a data grid is often large in scale, heterogeneous in its participating resources, and complicated in its architecture and network topology. Hence, we propose a topology reduction approach, called the Tree-to-Star Reduction method (TSR for short), with which we can reduce a complicated tree topology to a star in a data grid without impacting the evaluation accuracy and correctness of data transmission delays for a co-allocation scheme. The delivery star simplifies the evaluation and analytical complexities of the network delay time of a co-allocation scheme. Nevertheless, the TSR is also applicable to reducing the complexity of a delivery tree in other distribution and parallel systems, e.g., a grid computing, cloud computing, and P2P system. |
Relation: | 計畫編號:NSC101-2221-E029-003-MY3 研究期間:2012-08~ 2013-07 |
Appears in Collections: | [資訊工程學系所] 國科會研究報告
|
Files in This Item:
File |
Size | Format | |
index.html | 0Kb | HTML | 427 | View/Open |
|
All items in THUIR are protected by copyright, with all rights reserved.
|