DC 欄位 |
值 |
語言 |
DC.contributor | 資訊工程學系在職專班 | zh_TW |
DC.creator | 呂美靜 | zh_TW |
DC.creator | Mei-Jing Lyu | en_US |
dc.date.accessioned | 2012-8-10T07:39:07Z | |
dc.date.available | 2012-8-10T07:39:07Z | |
dc.date.issued | 2012 | |
dc.identifier.uri | http://ir.lib.ncu.edu.tw:88/thesis/view_etd.asp?URN=995302027 | |
dc.contributor.department | 資訊工程學系在職專班 | zh_TW |
DC.description | 國立中央大學 | zh_TW |
DC.description | National Central University | en_US |
dc.description.abstract | 由於高速的網際網路發展及普遍化與硬體效能和行動裝置的快
速運算需求提升下,網際網路成為未來資訊產業的主流,而雲端運算
的想法也因此順應而生,近年來更是各大企業與學術界關注的焦點。
在一個傳統的雲端資料中心,所有的資料都存儲在一個集中式的檔案
系統,如存儲區域網絡(SAN)檔案系統,並在許多電腦集群(computer
clusters)組成一個大的計算池執行計算任務。每當使用者請求計算
資源,從集中式檔案系統獲得所需的虛擬機圖像(VM image),然後
在計算池裡初始化虛擬機。當請求量多的時候,在此過程中可能產生
效能瓶頸,導致整個資料中心的效能降低,因此使用 Cache 的方法就
變得很重要。在這篇論文中,我們提出一些以統計資料或實驗參數來
訂定的 Cache 的解決策略,並設計了不同的虛擬機請求的情境,利用
cloudsim 雲模擬器去評估在不同情況下不同的 Cache 策略的效能如
何。最後找到一個最佳的 Cache 策略與容量,來達到減緩效能瓶頸與
提升整個資料中心的效能,並節省企業在 Cache 容量的花費成本。
| zh_TW |
dc.description.abstract | The Internet has become the foundation of the information
technology industry, and realized the cloud computing paradigm that
provides computing power to users in a pay-as-you-go manner. In a
typical cloud data center, all data are stored in a centralized file system
such as a storage-area-network (SAN) file system, and computation tasks
are executed in a large computing pool consisting of many computer
clusters. Whenever a user requests for a computation resource, the
management system of the data center should obtain the desired virtual
machine (VM) image from the centralized file system and then instantiate
the VM on the computing pool for the user. This procedure results in a
possible performance bottleneck, and may cause performance degradation
of the whole data center. In this paper, we have investigated several
caching strategies to see if they can alleviate the data transfer bottleneck
between the centralized file system and the computing pool in a data
center. We have designed different VM requests scenarios, and used a
cloud simulator, cloudsim, to evaluate the performance of different
caching strategies under different scenarios. Finally, we pointed out the
better caching strategy that can be incorporated in the design of a large
data center.
| en_US |
DC.subject | IaaS | zh_TW |
DC.subject | 雲端運算 | zh_TW |
DC.subject | 緩存策略 | zh_TW |
DC.subject | 資料中心 | zh_TW |
DC.subject | 資料傳輸瓶頸 | zh_TW |
DC.subject | data center | en_US |
DC.subject | Infrastructure-as-a-service | en_US |
DC.subject | data transfer bottleneck | en_US |
DC.subject | cloud computing | en_US |
DC.subject | caching strategy | en_US |
DC.title | 利用 Cache 改善雲端虛擬機器啟動之研究 | zh_TW |
dc.language.iso | zh-TW | zh-TW |
DC.title | Using Cache for Efficient Virtual MachineDeployment on Cloud Environments | en_US |
DC.type | 博碩士論文 | zh_TW |
DC.type | thesis | en_US |
DC.publisher | National Central University | en_US |