基于萤火虫群优化的虚拟机放置方法

doi:10.11959/j.issn.1000-0801.2022060

摘要/Abstract

摘要：

利用虚拟机放置策略对云数据中心的物理资源利用效率进行优化十分必要。提出了基于萤火虫群优化的虚拟机放置（glowworm swarm optimization based VM placement，Gso-wmp）方法。GSO-VMP方法将物理主机的处理器使用效率表示为荧光素值，当一个虚拟机被放置到物理主机上时，该物理主机的荧光素值都要进行更新；能够在局部径向范围内搜索到更多的可用物理主机，完成虚拟机放置，减少了虚拟机的迁移次数，从而间接地节省了物理主机的能量消耗。使用CloudSim作为GSO-VMP的仿真环境进行仿真，实验结果表明，GSO-VMP方法使得云数据中心的能耗降低、多维物理资源利用率提高。

关键词: 智能计算, 萤火虫群优化, 虚拟机放置, 云数据中心, 低能量消耗

Abstract:

In a cloud data center, one of the most important problems is using novel virtual machine placement strategy to promote the physical resource utilization.An approach of glowworm swarm optimization based virtual machine placement for cloud data centers called GSO-VMP was proposed.In the virtual placement, GSO algorithm was used to find a near-optimal solution.Each physical host had a luciferin value which represented the available CPU utilization.Whenever a VM was placed to a physical host, luciferin value of this physical host was updated.GSO-VMP algorithm could search the more available physical host within local range and thus the virtual migration numbers had been decreased and low energy consumption had been obtained.GSO-VMP had been evaluated using CloudSim with real-world workload data.The experimental results show that GSO-VMP has good performance in resource wastage and energy consumption.

Key words: intelligent computing, glowworm swarm optimization, virtual machine placement, cloud data center, low energy consumption

中图分类号:

TP393

徐胜超, 熊茂华, 周天绮. 基于萤火虫群优化的虚拟机放置方法[J]. 电信科学, 2022, 38(3): 172-182.

Shengchao XU, Maohua XIONG, Tianqi ZHOU. Approach of glowworm swarm optimization based virtual machine placement[J]. Telecommunications Science, 2022, 38(3): 172-182.

图/表 11

图1

图2

表1

表2

不同优化策略性能分析相关的参数设置"

算法	参数	具体值
GSO-VM	p荧光素衰减系数(0<p<1)	0.4
	物理主机 j的局部径向范围的最大传感范围ω_s	0.6
	相邻区域的变化率 θ	0.08
	最佳的邻居的数目k_t	5
	移动的步长尺寸s	0.03
	荧光素初始值 $l_{0}$	0.05
FPA^[4]	种群大小	50, 100, 150, 200
	标准函数gamma	1.5
	随机步长 L	[0, 1]
	转换概率 p	0.9
	最大迭代次数	1 000
GA^[22]	种群大小	50, 100, 150, 200
	交叉率	0.5
	变异率	0.01
	最大迭代次数	1 000
ACO^[29]	蚁群算法的蚂蚁个数	50, 100, 150, 200
	挥发因子 p	0.4
	信息素追踪权重α	0.3
	启发式信息权重β	1
	信息素更新常量 Q	100
	最大迭代次数	1 000

表2

表3

表4

图3

图4

图5

图6

表5

参考文献 30

[1]	陈双喜, 赵若琰, 刘会 ,等. 基于 KVM 的虚拟机 Post-Copy动态迁移算法稳定性优化[J]. 电信科学, 2021,37(7): 57-66.
	CHEN S X , ZHAO R Y , LIU H ,et al. Stability optimization of dynamic migration algorithm for Post-Copy of virtual machine based on KVM[J]. Telecommunications Science, 2021,37(7): 57-66.
[2]	黄丹池, 何震苇, 严丽云 ,等. Kubernetes容器云平台多租户方案研究与设计[J]. 电信科学, 2020,36(9): 102-111.
	HUANG D C , HE Z W , YAN L Y ,et al. Research and design of multi-tenant scheme for Kubernetes container cloud platform[J]. Telecommunications Science, 2020,36(9): 102-111.
[3]	SHI T , MA H , CHEN G . Energy-aware container consolidation based on PSO in cloud data centers[C]// Proceedings of 2018 IEEE Congress on Evolutionary Computation. Piscataway:IEEE Press, 2018: 1-8.
[4]	USMAN M J , ISMAIL A S , CHIZARI H ,et al. Energy-efficient virtual machine allocation technique using flower pollination algorithm in cloud datacenter:a panacea to green computing[J]. Journal of Bionic Engineering, 2019,16(2): 354-366.
[5]	ARIANYAN E , TAHERI H , KHOSHDEL V . Novel fuzzy multi objective DVFS-aware consolidation heuristics for energy and SLA efficient resource management in cloud data centers[J]. Journal of Network and Computer Applications, 2017(78): 43-61.
[6]	KAAOUACHE M A , BOUAMAMA S . Solving Bin packing problem with a hybrid genetic algorithm for VM placement in cloud[J]. Procedia Computer Science, 2015,60(1): 1061-1069.
[7]	BELOGLAZOV A , BUYYA R . Optimal online deterministic algorithms and adaptive heuristics for energy and performance efficient dynamic consolidation of virtual machines in cloud data centers[J]. Concurrency and Computation:Practice and Experience, 2012,24(13): 1397-1420.
[8]	WANG J V , CHENG C T , TSE C K . A power and thermal-aware virtual machine allocation mechanism for cloud data centers[C]// Proceedings of 2015 IEEE International Conference on Communication Workshop. Piscataway:IEEE Press, 2015: 2850-2855.
[9]	刘开南 . 云数据中心基于遗传算法的虚拟机迁移模型[J]. 计算机应用研究, 2020,37(4): 1115-1118.
	LIU K N . Virtual machine migration model in cloud data centers based on genetic algorithm[J]. Application Research of Computers, 2020,37(4): 1115-1118.
[10]	徐胜超 . 利用遗传算法完成虚拟机放置策略的优化[J]. 计算机与现代化, 2020(12): 25-31,42.
	XU S C . Using genetic algorithm for virtual machine placement optimization[J]. Computer and Modernization, 2020(12): 25-31,42.
[11]	徐胜超 . 贪心算法优化云数据中心的虚拟机分配策略[J]. 计算机系统应用, 2021,30(3): 134-141.
	XU S C . Greedy algorithms optimized virtual machine allocation for cloud data centers[J]. Computer Systems ＆ Applications, 2021,30(3): 134-141.
[12]	XIONG A P , XU C X . Energy efficient multiresource allocation of virtual machine based on PSO in cloud data center[J]. Mathematical Problems in Engineering, 2014:816518.
[13]	徐胜超 . 一种新的蚁群算法优化的虚拟机放置策略[J]. 计算机测量与控制, 2021,29(5): 235-240.
	XU S C . A new ant colony algorithm optimized virtual machine placement strategy[J]. Computer Measurement ＆ Control, 2021,29(5): 235-240.
[14]	陈艳, 周天绮, 徐胜超 . 利用蚁群算法完成虚拟机放置的优化[J]. 计算机工程与设计, 2021,42(5): 1229-1234.
	CHEN Y , ZHOU T Q , XU S C . ACO-VMP:using ant colony optimization algorithm for virtual machine placement[J]. Computer Engineering and Design, 2021,42(5): 1229-1234.
[15]	DUGGAN M , FLESK K , DUGGAN J ,et al. A reinforcement learning approach for dynamic selection of virtual machines in cloud data centres[C]// Proceedings of 2016 Sixth International Conference on Innovative Computing Technology (INTECH). Piscataway:IEEE Press, 2016: 92-97.
[16]	戴娇, 张明新, 孙昊 ,等. 花朵授粉算法的优化[J]. 计算机工程与设计, 2017,38(6): 1503-1509.
	DAI J , ZHANG M X , SUN H ,et al. Optimization of flower pollination algorithm[J]. Computer Engineering and Design, 2017,38(6): 1503-1509.
[17]	LUO J P , LI X , CHEN M R . Hybrid shuffled frog leaping algorithm for energy-efficient dynamic consolidation of virtual machines in cloud data centers[J]. Expert Systems With Applications, 2014,41(13): 5804-5816.
[18]	WANG J V , FOK K Y , CHENG C T ,et al. A stable matching-based virtual machine allocation mechanism for cloud data centers[C]// Proceedings of 2016 IEEE World Congress on Services. Piscataway:IEEE Press, 2016: 103-106.
[19]	WOOD T , SHENOY P , VENKATARAMANI A ,et al. Sandpiper:black-box and gray-box resource management for virtual machines[J]. Computer Networks, 2009,53(17): 2923-2938.
[20]	MISHRA M , SAHOO A . On theory of VM placement:anomalies in existing methodologies and their mitigation using a novel vector based approach[C]// Proceedings of 2011 IEEE 4th International Conference on Cloud Computing. Piscataway:IEEE Press, 2011: 275-282.
[21]	JOSEPH C T , CHANDRASEKARAN K , CYRIAC R . A novel family genetic approach for virtual machine allocation[J]. Procedia Computer Science, 2015(46): 558-565.
[22]	VASUDEVAN M , TIAN Y C , TANG M L ,et al. Energy-efficient application assignment in profile-based data center management through a repairing genetic algorithm[J]. Applied Soft Computing, 2018(67): 399-408.
[23]	LIU X F , ZHAN Z H , DENG J D ,et al. An energy efficient ant colony system for virtual machine placement in cloud computing[J]. IEEE Transactions on Evolutionary Computation, 2018,22(1): 113-128.
[24]	ALBOANEEN D A , TIANFIELD H , ZHANG Y . Glowworm swarm optimisation algorithm for virtual machine placement in cloud computing[C]// Proceedings of 2016 Intl IEEE Conferences on Ubiquitous Intelligence ＆ Computing,Advanced and Trusted Computing,Scalable Computing and Communications,Cloud and Big Data Computing,Internet of People,and Smart World Congress. Piscataway:IEEE Press, 2016: 808-814.
[25]	ZHOU Z , HU Z G , LI K Q . Virtual machine placement algorithm for both energy-awareness and SLA violation reduction in cloud data centers[J]. Scientific Programming, 2016:5612039.
[26]	JAMIL M , YANG X S . A literature survey of benchmark functions for global optimisation problems[J]. International Journal of Mathematical Modelling and Numerical Optimisation, 2013,4(2): 150.
[27]	WANG R , ZHOU Y Q . Flower pollination algorithm with dimension by dimension improvement[J]. Mathematical Problems in Engineering, 2014:481791.
[28]	LIN W W , XU S Y , HE L G ,et al. Multi-resource scheduling and power simulation for cloud computing[J]. Information Sciences, 2017,397/398: 168-186.
[29]	FARAHNAKIAN F , ASHRAF A , PAHIKKALA T ,et al. Using ant colony system to consolidate VMs for green cloud computing[J]. IEEE Transactions on Services Computing, 2015,8(2): 187-198.
[30]	SPEC. Benchmarks,standard performance evaluation corporation[S]. 2021.

云数据中心	参数	具体值
物理主机	内存	8 GB
	磁盘	1 TB
	操作系统	Linux
	网络带宽	1 Tbit/s
	处理器	Intel CoreTM i7
	虚拟机管理器	Xen
	CPU能量消耗模型	SpecPower

时间	虚拟机个数/个	物理主机个数/个	运行时间/h
周一	1 000～5 000	800	24
周二	1 000～5 000	800	24
周三	1 000～5 000	800	24
周四	1 000～5 000	800	24
周五	1 000～5 000	800	24

名称	处理器个数/个	内存需求/GB	磁盘需求/GB
类型1微型	2	3.75	2×8
类型2小型	4	7.5	2×16
类型3中型	8	15	2×40
类型4较大	16	30	2×80
类型5大型	32	60	2×160

虚拟机放置策略	资源利用	能量消耗/kw·h	活动物理主机数/个	虚拟机资源请求数/个
GSO-VMP	91.4%	100.3	450	2 000
GA	65.2%	125	736	2 000
ACO	71.5%	129	712	2 000
FPA	80%	111	606	2 000