通信学报
通信學報
통신학보
JOURNAL OF CHINA INSTITUTE OF COMMUNICATIONS
2014年
1期
72-81
,共10页
董健康%王洪波%李阳阳%程时端
董健康%王洪波%李暘暘%程時耑
동건강%왕홍파%리양양%정시단
IaaS%虚拟机放置%网络性能%能源效率
IaaS%虛擬機放置%網絡性能%能源效率
IaaS%허의궤방치%망락성능%능원효솔
IaaS%virtual machine placement%network performance%energy efficient
现在的虚拟机放置研究大多集中在物理服务器能源能耗或网络设备能耗的优化,然而随着这些资源的过度聚合,有可能会带来应用性能的下降。提出了一种虚拟机放置方案,主要有2个目的:最小化激活物理机和网络设备的个数来减少数据中心能源消耗;最小化最大链路利用率来改善网络性能。此方案在优化网络性能的同时,减少物理服务器和网络设备的能耗,使得能源效率与网络性能达到平衡。设计了一种新的二阶段启发式算法来求解,首先,利用基于最小割的层次聚类算法与最佳适应算法相结合来优化能源效率,然后,利用局部搜索算法再次优化虚拟机位置来最小化最大链路利用率。仿真实验结果表明,所提方案取得了良好的效果。
現在的虛擬機放置研究大多集中在物理服務器能源能耗或網絡設備能耗的優化,然而隨著這些資源的過度聚閤,有可能會帶來應用性能的下降。提齣瞭一種虛擬機放置方案,主要有2箇目的:最小化激活物理機和網絡設備的箇數來減少數據中心能源消耗;最小化最大鏈路利用率來改善網絡性能。此方案在優化網絡性能的同時,減少物理服務器和網絡設備的能耗,使得能源效率與網絡性能達到平衡。設計瞭一種新的二階段啟髮式算法來求解,首先,利用基于最小割的層次聚類算法與最佳適應算法相結閤來優化能源效率,然後,利用跼部搜索算法再次優化虛擬機位置來最小化最大鏈路利用率。倣真實驗結果錶明,所提方案取得瞭良好的效果。
현재적허의궤방치연구대다집중재물리복무기능원능모혹망락설비능모적우화,연이수착저사자원적과도취합,유가능회대래응용성능적하강。제출료일충허의궤방치방안,주요유2개목적:최소화격활물리궤화망락설비적개수래감소수거중심능원소모;최소화최대련로이용솔래개선망락성능。차방안재우화망락성능적동시,감소물리복무기화망락설비적능모,사득능원효솔여망락성능체도평형。설계료일충신적이계단계발식산법래구해,수선,이용기우최소할적층차취류산법여최가괄응산법상결합래우화능원효솔,연후,이용국부수색산법재차우화허의궤위치래최소화최대련로이용솔。방진실험결과표명,소제방안취득료량호적효과。
The existing virtual machine(VM) placement schemes mostly reduce energy consumption by optimizing utili-zation of physical server or network element. However, the aggressive consolidation of these resources may lead to net-work performance degradation. In view of this, a VM placement scheme was proposed to achieve two objectives. One is to minimize the number of activating physical machines and network elements to reduce the energy consumption, and the other is to minimize the maximum link utilization to improve the network performance. This scheme is able to reduce the energy consumption caused by physical servers and network equipment while optimizing the network performance, making a trade off between energy efficiency and network performance. A novel two-stage heuristic algorithm for a solu-tion was designed. Firstly, the hierarchical clustering algorithm based on minimum cut and best fit algorithm was used to optimize energy efficiency, and then, local search algorithm was used to minimize the maximum link utilization. The si-mulations show that this solution achieves good results.