本文關(guān)鍵詞：云中多租戶數(shù)據(jù)存儲系統(tǒng)的彈性控制問題研究　出處：《山東大學(xué)》2014年碩士論文　論文類型：學(xué)位論文

  更多相關(guān)文章： 多租戶 數(shù)據(jù)存儲 彈性 性能模型

【摘要】：云計算的出現(xiàn),為大型互聯(lián)網(wǎng)應(yīng)用提供新的交付模型,以其超大規(guī)模、高可擴(kuò)展性、按需服務(wù)等特點為下一代應(yīng)用開發(fā)帶來新的愿景。云中資源“無限性”以及pay-per-use的定價模式,使得應(yīng)用能夠借助云計算彈性特征應(yīng)對負(fù)載變化,達(dá)到傳統(tǒng)應(yīng)用無法獲得的彈性。云計算彈性是指在負(fù)載高峰期時申請更多資源以滿足性能要求,負(fù)載降低時合并空閑資源以降低整體資源開銷的能力,這是云計算主要特征之一。云中大型應(yīng)用因規(guī)模和復(fù)雜度增加,一般采用多層設(shè)計架構(gòu),而大部分應(yīng)用是數(shù)據(jù)驅(qū)動的,數(shù)據(jù)層面的彈性擴(kuò)展對整個系統(tǒng)擴(kuò)展至關(guān)重要。云計算環(huán)境數(shù)據(jù)存儲重點考慮可擴(kuò)展性,這相對于傳統(tǒng)關(guān)系型數(shù)據(jù)庫有諸多不同,由于數(shù)據(jù)存儲固有的有狀態(tài)性,通常獨占服務(wù)器,數(shù)據(jù)的管理要考慮到分布式事務(wù)特性、負(fù)載均衡的自適應(yīng)性以及靈活的副本一致性策略等問題,這使得數(shù)據(jù)層面的彈性控制難度增大。當(dāng)前,以多租戶為主要特征的SaaS應(yīng)用是最常見的云服務(wù),被越來越多的廠商所認(rèn)可,其兩大主要特征就是：單實例多租戶和系統(tǒng)能夠彈性適應(yīng)多租戶需求。多租戶的特征使得云中數(shù)據(jù)存儲系統(tǒng)的彈性控制更加復(fù)雜,面臨諸多挑戰(zhàn)： 1)數(shù)據(jù)彈性控制要在保證多租戶的數(shù)據(jù)共享存儲和滿足不同租戶數(shù)據(jù)的性能要求下進(jìn)行。云提供商一般為了減少整體資源的開銷,租戶的數(shù)據(jù)會共享存儲在分布式節(jié)點上。不同租戶對于數(shù)據(jù)訪問的性能要求是不同的,具體體現(xiàn)在SLA中,本文關(guān)注數(shù)據(jù)訪問時間相關(guān)的指標(biāo)。在進(jìn)行數(shù)據(jù)彈性控制時要保證以上兩個條件,這增加了彈性控制的難度。 2)為了實現(xiàn)多租戶存儲系統(tǒng)的彈性控制,我們需要能夠提出控制策略,決定哪些數(shù)據(jù)要進(jìn)行移動,往哪里去移動。大型應(yīng)用的負(fù)載波動大,自適應(yīng)的負(fù)載均衡要求能夠在負(fù)載高峰時scale-out,負(fù)載低峰時scale-down。這通過數(shù)據(jù)移動(比如,partitioned或coalesced)實現(xiàn),但是頻繁的移動有可能為已經(jīng)過載的系統(tǒng)帶來新的性能問題。 3)要有一種負(fù)載預(yù)測機(jī)制使得數(shù)據(jù)擴(kuò)展策略能夠提前制定。云中資源供應(yīng)要經(jīng)歷啟動延遲,這就要求必須盡早應(yīng)對負(fù)載變化做出控制策略。監(jiān)測和控制服務(wù)水平協(xié)議(SLA)中的性能指標(biāo)容易受環(huán)境噪聲影響,不易準(zhǔn)確測量,而且不同租戶不同SLA中性能要求相對于相同性能要求的問題更加復(fù)雜。 針對上述問題,我們提出一種基于MPC控制思想實現(xiàn)的AdaptScala系統(tǒng),該系統(tǒng)能夠監(jiān)測不同租戶數(shù)據(jù)訪問的負(fù)載情況,并基于一次指數(shù)平滑方法進(jìn)行負(fù)載預(yù)測；結(jié)合不同租戶SLA中對數(shù)據(jù)訪問響應(yīng)時間的性能要求,構(gòu)建多租戶數(shù)據(jù)存儲的性能模型用于判定每個服務(wù)器能否滿足不同租戶對數(shù)據(jù)訪問的性能要求；最后給出一種數(shù)據(jù)彈性控制策略生成算法,用于計算最終的數(shù)據(jù)存儲調(diào)整策略,其目標(biāo)是對系統(tǒng)整體性能影響較小,并且使整體資源開銷最小。 本文采用Berkeley DB作為云中key/value類型的數(shù)據(jù)存儲引擎,使用開源軟件構(gòu)建實現(xiàn)AdaptScala系統(tǒng),模擬多租戶的數(shù)據(jù)訪問并采集性能數(shù)據(jù)構(gòu)建多租戶數(shù)據(jù)存儲模型,進(jìn)行多次測試驗證該系統(tǒng)能夠滿足提出的彈性控制目標(biāo)。
[Abstract]:Cloud computing is one of the main characteristics of cloud computing . The cloud computing elasticity refers to the ability to apply more resources to meet performance requirements and to reduce overall resource overhead by means of cloud computing elastic characteristics . 1 ) The data elastic control is carried out under the performance requirement of ensuring multi - tenant data sharing and meeting the performance requirements of different tenant data . Generally , cloud providers share the data stored on the distributed nodes in order to reduce the overhead of the whole resources . The performance requirements of different tenants for data access are different , and the indexes of the data access time are expressed in the SLA . In the data elastic control , the above two conditions are ensured , which increases the difficulty of elastic control . 2 ) In order to realize the elastic control of multi - tenant storage system , we need to be able to put forward control strategy , decide which data is to be moved , where to move . The load fluctuation of large - scale application is big , and the adaptive load balancing requirement can scale - out when the load peak , scale - down when the load is low . This can be realized by data movement ( for example ) , but the frequent movement has the potential to bring new performance problem for the already overloaded system . 3 ) There is a load forecasting mechanism that enables data scaling strategies to be developed in advance . The resource supply in the cloud is subject to start - up delay , which requires that control strategies should be taken as soon as possible . The performance indicators in the monitoring and control service level agreement ( SLA ) are easily affected by environmental noise , difficult to accurately measure , and performance requirements of different tenants in different SLAs are more complex with respect to the same performance requirements . In view of the above - mentioned problems , we propose a system based on MPC control . The system can monitor the load situation of different tenant data access and load forecast based on the exponential smoothing method . The performance model of multi - tenant data storage is used to determine whether each server can meet the performance requirement of different tenants for data access . Finally , a data elasticity control strategy generation algorithm is presented to calculate the final data storage adjustment strategy . The goal is to influence the overall performance of the system and minimize the overall resource overhead .

鏈枃閲囩敤Berkeley DB浣滀負(fù)浜戜腑key/value綾誨瀷鐨勬暟鎹瓨鍌ㄥ紩鎿，

本文編號：1360920