多实例云计算资源市场下超额预订决策方法

doi:10.11772/j.issn.1001-9081.2016.01.0113

计算机应用 ›› 2016, Vol. 36 ›› Issue (1): 113-116.DOI: 10.11772/j.issn.1001-9081.2016.01.0113

多实例云计算资源市场下超额预订决策方法

陈冬林¹, 姚梦迪¹, 邓国华^1,2

1. 武汉理工大学电子商务与智能服务研究中心, 武汉 430070;
2. 江汉大学商学院, 武汉 430056

收稿日期:2015-07-22 修回日期:2015-09-24 出版日期:2016-01-10 发布日期:2016-01-09
通讯作者: 姚梦迪(1991-),女,湖北安陆人,博士研究生,主要研究方向:云计算、智能推荐
作者简介:陈冬林(1970-),男,湖北安陆人,教授,博士,主要研究方向:云计算、商务智能、互联网经济;邓国华(1984-),男,湖北武汉人,博士研究生,主要研究方向:云计算、智能推荐、数据分析。
基金资助:
国家自然科学基金资助项目(71172043);教育部回国留学人员科研启动基金资助项目(2013-693);武汉理工大学研究生自由探索项目(2015-zy-126)。

Overbooking decision-making method of multiple instances under cloud computing resource market

CHEN Donglin¹, YAO Mengdi¹, DENG Guohua^1,2

1. Research Center of E-commerce and Intelligence Service, Wuhan University of Technology, Wuhan Hubei 430070, China;
2. School of Business, Jianghan University, Wuhan Hubei 430056, China

Received:2015-07-22 Revised:2015-09-24 Online:2016-01-10 Published:2016-01-09
Supported by:
This work is partially supported by the Surface Program of National Natural Science Foundation of China (41471329).

摘要/Abstract

摘要： 针对现有云供应商数据中心负载率低、云用户需求不确定及多样性的问题,为提高云供应商平均利润,建立了不确定需求下的多实例类型云服务超额预订模型。该模型结合实际云计算资源市场下超额预订对于云供应商负载均衡及云服务等级协议(SLA)的影响,给出超额预订的多重约束条件,提出了各实例类型数量最优分配策略。实验结果表明,采用该模型在预约未使用概率为0.25时,云供应商利润较高,数据中心负载率达到78%,最终确定了各实例类型的最优分配数量。

关键词: 不确定需求, 云计算资源市场, 多实例, 超额预订

Abstract: Considering the problems of low load rate of data centers in cloud providers, uncertainty and variety of cloud user demand; in order to improve the average profit of the cloud providers, an overbooking model of multiple instances under uncertain demand was established. The proposed model combined the influences of overbooking for cloud data center load balancing and Service Level Agreement (SLA) under the actual cloud computing resource market, multi-constraint of overbooking was provided, then the optimal allocation policy of each instance type was put forward. The simulation results show that when the unused rate of reservation is 0.25, the average profit is relatively high, the load rate of data center is 78%, finally the optimal allocation of each instance type is determined.

Key words: uncertain demand, cloud computing resource market, multi-instance, overbooking

中图分类号:

TP393

陈冬林, 姚梦迪, 邓国华. 多实例云计算资源市场下超额预订决策方法[J]. 计算机应用, 2016, 36(1): 113-116.

CHEN Donglin, YAO Mengdi, DENG Guohua. Overbooking decision-making method of multiple instances under cloud computing resource market[J]. Journal of Computer Applications, 2016, 36(1): 113-116.

参考文献

[1] 冯登国,张敏,张妍,等.云计算安全研究[J].软件学报,2011,22(1):71-83.(FENG D G, ZHANG M, ZHANG Y, et al. Study on cloud computing security [J]. Journal of software, 2011,22(1):71-83.)
[2] 屈喜龙,肖鹏,白泉.云环境中基于弹性预留机制的虚拟资源供给策略[J].系统工程理论与实践,2015,35(6):1573-1581.(QU X L, XIAO P, BAI Q. Virtual resource provision policy based on elastic reservation mechanism in cloud environment [J]. System engineering theory and practice, 2015, 35(6): 1573-1581.)
[3] HOUSEHOLDER R, ARNOLD S, GREEN R. Simulating the effects of cloud-based oversubscription on datacenter revenues and performance in single and multi-class service levels [C]// Proceedings of the 2014 IEEE 7th International Conference on Cloud Computing (CLOUD). Piscataway, NJ: IEEE, 2014: 562-569.
[4] 朱金福,刘玮,高强.航空客运舱位控制和超售综合静态建模研究[J].中国管理科学,2006,14(5):68-72.(ZHU J F, LIU W, GAO Q. Integrated static modeling of airline seat inventory control and overbooking [J]. Chinese journal of management science, 2006, 14(5): 68-72.)
[5] 赵昊天,贾传亮,宋砚秋,等.不确定需求下航空超售问题的鲁棒优化研究[J].中国管理科学,2013,21(11):98-101.(ZHAO H T, JIA C L, SONG Y Q, et al. Research on uncertain demands of airline overbooking model based on adjustable robust optimization [J]. Chinese journal of management science, 2013, 21(11): 98-101.)
[6] 周蔷,刘长有.基于随机特性的航空机票动态超售模型[J].系统工程理论与实践,2014,34(3):717-722.(ZHOU Q, LIU C Y. Based on the stochastic characteristics of the airplane ticket dynamic overbooking model [J]. System engineering—theory & practice, 2014, 34(3): 717-722.)
[7] 徐丽萍,李金林,雷俊丽,等.基于超订的民航收益管理单航段舱位控制模型比较研究[J].系统工程理论实践,2014,34(1):129-137.(XU L P, LI J L, LEI J L, et al. Comparative analysis on one-leg airline capacity control models with overbooking [J]. System engineering—theory & practice, 2014, 34(1): 129-137.)
[8] KIM S, KIM H, LEE J, et al. Group-based memory oversubscription for virtualized clouds [J]. Journal of parallel and distributed computing, 2014, 74(4): 2241-2256.
[9] RACHEL H, SCOTT A, ROBERT G. Simulating the effects of cloud-based oversubscription on datacenter revenues and performance in single and multi-class service levels [C]// Proceedings of the 2014 IEEE International Conference on Cloud Computing. Piscataway, NJ: IEEE, 2014: 562-596.
[10] PVSCHEL T, SCHRYEN G, HRISTOVA D, et al. Revenue management for cloud computing providers: decision models for service admission control under non-probabilistic uncertainty [J]. European journal of operational research, 2015, 244(2): 637-647..
[11] Amazon EC2 instance [EB/OL]. [2014-02-05]. http://aws.amazon.com/cn/ec2/instance-types/.
[12] KHEMKA B, FRIESE R, PASRICHA S, et al. Utility maximizing dynamic resource management in an oversubscribed energy-constrained heterogeneous computing system [J]. Sustainable computing: informatics and systems, 2014, 5: 14-30.
[13] MACÍAS M, GUITART J. SLA negotiation and enforcement policies for revenue maximization and client classification in cloud providers [J]. Future generation computer systems, 2014, 41: 19-31.

多实例云计算资源市场下超额预订决策方法

Overbooking decision-making method of multiple instances under cloud computing resource market

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 3

编辑推荐

Metrics

[1]	彭爽, 彭晓明. 基于高效多示例学习的目标跟踪[J]. 计算机应用, 2015, 35(2): 466-469.
[2]	梁爱南李长云黄贤明 . 多实例工作流模式的π演算形式化[J]. 计算机应用, 2007, 27(1): 219-220.
[3]	许甸李建华刘星沙夏媛 . 支持多实例模式的动态工作流研究[J]. 计算机应用, 2006, 26(12): 3033-3036.