一個在 O p e n S t a c k 平台 進行 混 合 式 自 動 擴 展 的 方 法

虛擬化技術及網路環境尚未成熟前，機房建置主要透過實體機器組成伺服器群組來提供計算服務。隨著近幾年虛擬化技術及網路環境發展迅速，許多雲端上的計算機房也開始由傳統實體機房轉型為虛擬化機房，並透過虛擬機提供計算服務。當虛擬化機房面臨突然快速增加之計算請求時，一般會採用額外新增伺服器或者虛擬機器來分擔計算負荷，而這種方式稱為虛擬叢集之擴展，反觀傳統機房對於上述狀況無法如此迅速因應。而在實務上，例如 OpenStack 的自動擴展機制，採用的方法是反應式自動擴展機制，也就是去檢查系統資源使用是否超過預定的數值(Threshold)後進行資源調整。在本論文研究中，我們以預測式自動擴展機制為基礎，透過歷史資料來預測未來工作負載，使得伺服器叢集可以為即將到來的工作負載，提早準備好資源因應。本論文並將統計學的預測方法實作到 OpenStack 上，與反應式自動擴展結合，提出混合式自動擴展機制，最後透過實驗結果驗證本論文提出之策略，能在系統面臨大量工作負載來臨時，對於使用者請求完成數及回應速度有著優異表現，並且不會對系統造成額外負擔。關鍵字: 自動擴展、OpenStack、負載平衡、反應式自動擴展、預測式自動擴展、虛擬機。

關鍵字

自動擴展；混合式

並列摘要

The use of virtualization technology has gradually changed the way a datacenter works in recent years. Nowadays the end-users of a datacenter do not access physical resources directly. Instead, they access virtualized resources, such as VMs and virtual clusters, on top of a pool of physical resources. This new computing paradigm provides the datacenter administrators a more flexible, scalable, manageable, and economical way for resource provisioning/sharing as prior study indicated. When a service on a VM encounters a massive amount of workload, it can scale faster than a non-virtualized datacenter, by dynamically turning on extra virtual/physical machines to share the workload. For example, OpenStack, an open source project for building a virtualized cloud platform, provides a reactive approach for auto-scaling. That is, it creates new VMs to share workload when the workload of a monitored VM exceeds a given workload threshold. The weakness of the mechanism is that, sometimes it is too late to handle unexpected workload surges and thus can decrease the quality of the services running on the VM. To this end, we purpose a new hybrid auto-scaling mechanism for auto-scaling. It relies on a predictive auto-scaling approach that predicts the upcoming workload by historical workloads. To prevent the case that the prediction result is not accurate enough, we also use the reactive auto-scaling mechanism provided by OpenStack, and integrate the two mechanisms as one. We have verified the performance of our approach via experiments, and the results show that, when a massive workload arrives, the proposed approach outperforms other approaches. In addition, the proposed approach does not incur much overhead as the experimental results show. Keywords: Auto-scaling, reactive auto-scaling, predictive auto-scaling, load balancing, load sharing, virtual machine

並列關鍵字

OpenStack ； auto-scaling ； hybrid

參考文獻

[1] Y. Jadeja, and K. Modi, “Cloud computing - concepts, architecture and

(ICCEET), 2012 International Conference on, 2012, pp. 877-880.

[2] A. S. Idris, N. Anuar, M. M. Misron, and F. H. M. Fauzi, “The readiness of

Cloud Computing: A case study in Politeknik Sultan Salahuddin Abdul Aziz

[4] F. B. Shaikh, and S. Haider, “Security threats in cloud computing,” in

國際替代計量

一個在 O p e n S t a c k 平台進行混合式自動擴展的方法

未授權

主題瀏覽