天堂国产午夜亚洲专区-少妇人妻综合久久蜜臀-国产成人户外露出视频在线-国产91传媒一区二区三区

私有云平臺服務能力檢測方法

發(fā)布時間:2018-10-25 20:10
【摘要】:針對現(xiàn)階段使用私有云評測方法缺失所導致搭建私有云供需不匹配的問題,提出了一種自適應、可擴展的私有云系統(tǒng)能力檢測方法,可對基礎設施即服務(IaaS)云計算系統(tǒng)能力進行評測。首先,通過私有云應用程序接口動態(tài)擴展虛擬機數(shù)量;然后,通過性能特征模型選取虛擬機所需配置的硬件信息和操作系統(tǒng)類別,并根據(jù)用戶的需求差異使用不同的負載模型,形成相應的模擬運行環(huán)境;最后,使用云計算服務等級協(xié)議(SLA)作為測試標準來衡量私有云服務能力。使用所提方法對Openstack開源私有云平臺的能力進行了評測,實驗結(jié)果表明,所提方法可以準確得出私有云服務能力,結(jié)果和用戶實測一致。該方法負載生成成本更低,測試效率更高。并將該方法與Openstack自帶的組件Rally進行對比,拓展性和負載動態(tài)模擬有很大的改進。
[Abstract]:Aiming at the mismatch between supply and demand of private cloud due to the lack of private cloud evaluation method at present, an adaptive and extensible method for detecting the capability of private cloud system is proposed. The capability of infrastructure as a service (IaaS) cloud computing system can be evaluated. Firstly, the number of virtual machines is dynamically extended through the private cloud application program interface. Then, the hardware information and operating system categories of virtual machine are selected through the performance feature model, and different load models are used according to the user's demand difference to form the corresponding simulation running environment. Finally, the private cloud service capability is measured using cloud service level protocol (SLA) as a test standard. The proposed method is used to evaluate the capability of Openstack open source private cloud platform. The experimental results show that the proposed method can accurately obtain the private cloud service capability, and the results are consistent with user measurements. The load generation cost of the method is lower and the test efficiency is higher. The method is compared with the component Rally of Openstack, and the expansibility and load dynamic simulation are greatly improved.
【作者單位】: 西北工業(yè)大學計算機學院;
【基金】:國家863計劃項目(2015AA015307) 國家自然科學基金資助項目(61472323) 中央高;究蒲袠I(yè)務費專項(3102015JSJ0009)~~
【分類號】:TP393.09


本文編號:2294714

資料下載
論文發(fā)表

本文鏈接:http://sikaile.net/guanlilunwen/ydhl/2294714.html


Copyright(c)文論論文網(wǎng)All Rights Reserved | 網(wǎng)站地圖 |

版權(quán)申明:資料由用戶93985***提供,本站僅收錄摘要或目錄,作者需要刪除請E-mail郵箱bigeng88@qq.com