欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
当前，云数据中心的能耗问题已成为业界关注的热点．已有研究工作大多致力于从技术角度降低数据中心的能耗，或在能耗与性能之间寻求一种最佳的折衷．云计算作为一种商业计算模式，已有研究成果很少考虑到云定价策略对能耗管理机制的影响．文中提出了基于动态定价策略的数据中心能耗成本优化方案．建立起服务价格和能耗成本的统一模型，通过研究两者之间的关系．协同优化服务价格与能耗成本，使数据中心的收益达到最优．鉴于数据中心规模庞大、承载任务繁重等特点，论文采用基于重载近似的大规模排队系统来对数据中心建模，根据不同数据中心问的服务需求量和电价差别，设计了多数据中心间的负载路由机制，旨在削减数据中心的整体能耗成本．针对单个数据中心，定义了双阈值策略以动态调节服务器的各种状态（0n／0if／Idle等），从而使数据中心能耗成本得到进一步优化．实验结果表明，论文提出的解决方案能够在满足用户Q0s需求的前提下，较好地优化数据中心能耗成本，同时使数据中心的收益达到最优．
英文摘要：
Minimizing the energy consumption is one of the most important study areas in green data center. In the past, most energy efficiency mechanisms proposed in the literature are to min- imize the total energy consumption or find a tradeoff between energy and perform. However, existing research on power management with little regard to the effect of pricing strategy. In this paper we propose a data center energy costs optimization mechanism based on a dynamic pricing strategy. By studying the relationship between two of them, we establish a unified model for the pricing and energy costs and achieve to the optimal benefits of the whole data centers. In view of the fact that the data center is large size and takes heavy traffic, this paper models the data center as a heavy traffic approximate large-scale queuing system. According to the differences of service demands and electricity prices, we design a load routing mechanism between multiple data centers to reduce the overall data center energy costs. For a single data center, a dual-threshold strategy is defined to dynamically adjust the various states of a single server （On/Off/Idle, etc. ）, so that to reduce the energy costs of the data centers. Experimental results show that solution proposed in this paper are able to meet the users＇ QoS requirements under the premise of better optimiza- tion of data center energy costs, while making the data center to achieve optimal benefits.
同期刊论文项目
同项目期刊论文
Optimization of Gateway Deployment with Load Balancing and Interference Minimization in Wireless Mes
Towards optimized scheduling for data-intensive scientific workflow in multiple datacenter environme
Scientific workflow scheduling in non-dedicated heterogeneous multicluster with advance reservations
期刊信息
位置：