English  |  正體中文  |  简体中文  |  Items with full text/Total items : 49287/83828 (59%)
Visitors : 7151614      Online Users : 85
RC Version 7.0 © Powered By DSPACE, MIT. Enhanced by NTU Library & TKU Library IR team.
Scope Tips:
  • please add "double quotation mark" for query phrases to get precise results
  • please goto advance search for comprehansive author search
  • Adv. Search
    HomeLoginUploadHelpAboutAdminister Goto mobile version
    Please use this identifier to cite or link to this item: http://tkuir.lib.tku.edu.tw:8080/dspace/handle/987654321/74707


    Title: 架構於雲端運算系統儲存之動態負載平衡機制
    Other Titles: Dynamic load balancing mechanism based on cloud computing storage system
    Authors: 林益新;Lin, Yih-Sin
    Contributors: 淡江大學電機工程學系碩士在職專班
    吳庭育
    Keywords: 雲端計算系統;負載平衡;P2P;Cloud system;load balance
    Date: 2011
    Issue Date: 2011-12-28 19:22:00 (UTC+8)
    Abstract: 由於網路的普及與便利性,讓目前的3C產品都開始具備網路功能以符合使用者的期望,因應這樣的趨勢雲端運算也就因應而生。對於儲存協定,雲端運算也有相當深入的定義;在儲存協定中,檔案的儲存與管理都是藉由第三方的實體儲存設備,並經由集中式的核心伺服器進行整合與分析,將整合的可用資源,依據使用者的需求進行資料傳輸與保存。
    依據雲端運算的架構,其涵蓋的範圍相當大,所轄的網域範圍相當多,在這種情境下,多數使用者會因應使用習慣及所擁有資源執行多種重複性儲存行為及存取高相似度資料。假設儲存設備中所讀寫的內容都有著高相似度與重複性,系統管理者將無法確保雲端運算系統中的每一個儲存節點(Peer)都能保持最佳狀態來進行資料整合的工作;並且當節點不斷擴增,其檔案數量則會倍數成長,屆時將造成系統管理者的工作負擔,也會增加硬體資源的浪費,逐漸失去雲端儲存系統的彈性及速度,只剩下龐大及重複的冗餘資料。
    本論文為了避免存取資料高度重複而產生的系統負擔,提出一種新式資料中心管理架構:索引名稱伺服器 (Index Name Server, INS),INS整合了重複資料比對機制以及存取節點選擇最佳化等相關機制來提升整體雲端資料的存取效能。本論文應用INS特有參數來動態監測IP索引資料及忙碌品質等參數,使整體負載達到平衡。避免資料傳輸時,發生長時間排隊或傳輸擁塞發生。經實驗的驗證可以有效提升整體傳輸效能達到20%~50%。
    Thanks to the popularity and convenience of networks, almost all 3C products nowadays can connect to the network in accordance with users’ expectation. For this trend, cloud computing takes shape. Cloud computing has defined the storage protocol thoroughly: data storage and management are executed by the physical storage devices of the third party. After the integration and analysis by the centralized server core, available resource and data is transmitted and stored according to user’s needs.
    Since the coverage and domain of cloud computing-based architecture are very extensive, according to users’ habits and available resources, most users save and access the data of high similarity. Supposing the data stored in the storage devices is of high duplication and similarity, the system administrator cannot guarantee that every node (peer) in the cloud computing system can maintain the best status for data integration. Moreover, when the number of nodes keeps increasing, the number of files will be multiplied, which results to not only the heavy workload of the system administrator, but also the waste of hardware resources. Finally, the flexibility and quickness of cloud computing system is gone, and what left in the system is big, duplicated and unnecessary data.
    Therefore, in order to avoid the system burden cased by data duplication, we propose a novel data center management structure: Index Name Server, INS, which integrates schemes, like the de-duplication and the optimized access point selection, to enhance the performance of the cloud computing system. In this paper, with the attempt to attain load balancing, we use several specific parameters of INS to dynamically monitor the parameters like IP information and busy quality index. In this way, the network congestion or the long waiting time can be avoided in data transmission. Our experimental result proves that our proposed method can achieve 20% to 50% improvement for transmission performance.
    Appears in Collections:[電機工程學系暨研究所] 學位論文

    Files in This Item:

    File SizeFormat
    index.html0KbHTML106View/Open

    All items in 機構典藏 are protected by copyright, with all rights reserved.


    DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library & TKU Library IR teams. Copyright ©   - Feedback