云计算环境中绿色服务级目标的分析、量化、建模及评价

Profiling,Quantifying,Modeling and Evaluating Green Service Level Objectives in Cloud Computing Environments

高能效性问题是开启绿云计算新纪元的主要障碍之一.特别是随着全球气候变化和对能量依赖程度的不断增加,绿色计算作为一种新的高端计算,受到了人们的广泛关注.为了实现云计算能耗的最小化,实现对绿色服务级目标的可量化和可管理,我们开展了关于云计算环境中绿色服务级目标的分析、量化、建模及评价的研究.在本文中,首先给出了绿云的定义.随后,通过借鉴服务级目标和绿色计算的相关理论,对实现绿色服务级目标的原理进行了系统的分析.之后,基于绿云的原理和方法,提出了一种多维能耗模型M2EC.理论和实验结果均表明M2EC模型在高能效方面具有明显优势,实现了云计算环境中能耗和服务级目标之间的均衡.

High energy efficiency issue is one of the major obstacles for opening up the new era of the long dreamed vision of computing as a green computing with large-scale cloud data centers. In particular, with the global climate change and the ever growing dependence on energy, green computing, as a high-end computing platform, has received extensive attention. To minimize en- ergy consumption and to achieve the quantifiability and manageability of the high green service level objectives for cloud computing environments, the analysis, quantification, modeling and evaluation of the green service level objectives for cloud computing environments are investigated. In this paper, the definition of green cloud is given and the principles for implementing high green service level objectives are systematically analyzed by referring to the service level objective and green computing theories. Based on the principles and methodology of green cloud, a multi-metric energy consumption model M2EC is put forward. Theoretical as well as experimental results conclusively demonstrate that the M2EC algorithm has high potential as it provides etllclent green enhancements and significant energy saving. It implements the trade-off between energy con- sumption and green-service level obiectives efficiently and effectively in cloud computing environ- ments.