標題: | 基於SLA之雲端資料中心負載平衡機制 An SLA-aware load balancing scheme for cloud datacenters |
作者: | 黎中誠 Li, Chung-Cheng 王國禎 Wang, Kuo-Chen 網路工程研究所 |
關鍵字: | 類神經網路;雲端計算;非集中式架構;負載平衡;服務水準協議;Artificial neural network;cloud computing;decentralized architecture;load balancing;service level agreement |
公開日期: | 2011 |
摘要: | 雲端計算這個名詞出現在2007年的第四季,它具有高度的可延展性和接近無限的 (如計算) 資源。如何讓擁有成千上萬的虛擬機器的大型資料中心達到負載平衡是雲端計算的重要議題之一。在本篇文章中,我們提出了一種新的非集中式負載平衡結構,稱為雙層非集中式負載平衡器 (tldlb )。這種非集中式負載平衡器擁有可延展性和高可用性的優點,有利於服務更多雲端使用者。除此之外,我們也提出一個稱為基於類神經網路之動態加權循環 (nn-dwrr) 的動態負載平衡演算法,它能有效地將大量的使用者請求分配到各個實際提供服務的虛擬機器上。在nn-dwrr演算法中,我們將監控虛擬機器所得的負載指標(CPU、記憶體、網路頻寬、硬碟存取等四項利用率)和類神經網路結合,以調整每台虛擬機器的服務權重。我們的nn-dwrr演算法可以利用類神經的預測和最佳化能力,有效減低服務水準協議 (SLA) 的違反率。實驗結果證明我們所提出的負載平衡演算法 (nn-dwrr) 在資源有限的情況下,其平均回應時間上比wrr快1.86倍,比capacity based快1.49倍,以及比ANN-based快1.21倍。我們的方法,在相同時間內能處理更多的使用者要求,而與其他負載平衡演算法相比,更適用於大型雲端資料中心。此外,tldlb 演算法可以適時啟動虛擬機器池中的虛擬機器來避免違反SLA。 Cloud computing appears at the fourth season, 2007. It has high scalability and nearly unlimited (e.g., computing) resources. One of the most important issues about cloud computing is how to achieve load balancing among thousands of virtual machines (VMs) in a large datacenter. In this paper, we propose a novel decentralized load balancing architecture, called tldlb (two-level decentralized load balancer). This distributed load balancer takes advantage of the decentralized architecture for providing scalability and high availability capabilities to service more cloud users. We also propose a neural network-based dynamic load balancing algorithm, called nn-dwrr (neural network-based dynamic weighted round-robin), to dispatch a large number of client requests to different VMs, which are actually providing services. In nn-dwrr, we combine of VM load metrics monitoring (CPU, memory, network bandwidth, disk I/O utilization) and neural network to adjust the weight of each VM. Our nn-dwrr algorithm can reduce SLA (service-level agreement) violations. Experimental results support that our proposed load balancing algorithm, nn-dwrr, can be applied to a large cloud datacenter, and it is 1.86 times faster than wrr, 1.49 times faster than capacity-based, and 1.21 times faster than ANN-based load balancing algorithms in terms of average response time in the limited resources. In addition, tldlb can avoid SLA violations via in-time activating VMs in the spare VM pool. |
URI: | http://140.113.39.130/cdrfb3/record/nctu/#GT079956539 http://hdl.handle.net/11536/50574 |
顯示於類別: | 畢業論文 |