基于压缩因子的宽度学习系统的虚拟机性能预测

邹伟东; 夏元清

doi:10.16383/j.aas.c190307

基于压缩因子的宽度学习系统的虚拟机性能预测

doi: 10.16383/j.aas.c190307

邹伟东^1,,
夏元清^1,

1.
北京理工大学自动化学院北京 100081

基金项目: 国家重点研发计划(2018YFB1003700), 国家自然科学基金(61836001)资助

详细信息

作者简介:
邹伟东：北京理工大学自动化学院博士后. 主要研究方向为宽度学习系统, 云数据中心优化调度管理. E-mail: zouweidong1985@163.com

夏元清：北京理工大学自动化学院教授. 主要研究方向为云控制, 云数据中心优化调度管理, 智能交通, 模型预测控制, 自抗扰控制, 飞行器控制和空天地一体化网络协同控制. 本文通信作者. E-mail: xia_yuanqing@bit.edu.cn

计量
- 文章访问数: 594
- HTML全文浏览量: 219
- PDF下载量: 137
- 被引次数: 0
出版历程
- 收稿日期: 2019-04-17
- 录用日期: 2019-07-30
- 网络出版日期: 2022-02-18
- 刊出日期: 2022-03-25

Virtual Machine Performance Prediction Using Broad Learning System Based on Compression Factor

ZOU Wei-Dong^1
,,
XIA Yuan-Qing^1
,

1.
School of Automation, Beijing Institute of Technology, Beijing 100081

Funds: Supported by National Key Research and Development Program of China (2018YFB1003700) and National Natural Science Foundation of China (61836001)

More Information

Author Bio:
ZOU Wei-Dong　Postdoctoral fellow at the School of Automation, Beijing Institute of Technology. His research interest covers broad learning system, cloud data center optimization scheduling and management

XIA Yuan-Qing　Professor at the School of Automation, Beijing Institute of Technology. His research interest covers cloud control, cloud data center optimization scheduling and management, intelligent transportation, model predictive control, active disturbance rejection control, and flight control and networked cooperative control for integration of space, air and earth. Corresponding author of this paper

摘要

摘要: 在基于基础设施即服务的云服务模式下, 精准的虚拟机性能预测, 对于用户在众多资源提供商之间进行虚拟机租用策略的制定具有十分重要的意义. 针对基于宽度学习系统(Broad learning system, BLS)的预测模型存在许多降低虚拟机性能预测准确性和效率的冗余节点, 通过引入压缩因子, 构建基于压缩因子的宽度学习系统, 使预测结果更逼近输出样本, 能够减少BLS的冗余特征节点与增强节点, 从而加快BLS的网络收敛速度, 提高BLS的泛化性能.
- 虚拟机性能预测 /
- 宽度学习系统 /
- 压缩因子 /
- 网络收敛速度 /
- 泛化性能
Abstract: In cloud service models which is based on IaaS, from the user＇s perspective, how to accurately predict performance of virtual machine is very important for making renting strategy of virtual machines between many physical servers. However, broad learning system (BLS) includes too many redundant feature nodes and enhancement nodes, resulting in decreased efficiency and accuracy of virtual machine performance prediction. Connecting compression factor to BLS, the paper builds intelligent prediction model of BLS based on compression factor (CF-BLS), and uses the model for predicting virtual machine performance.
- Performance prediction of virtual machine /
- broad learning system (BLS) /
- compression factor /
- convergence rate of network /
- generalization performance

HTML全文

图 1 Combined Cycle Power Plant数据集对CF-BLS与BLS算法的RMSE和PCC曲线

Fig. 1 Curves for RMSE and PCC of Combined Cycle Power Plant dataset based on CF-BLS and BLS

下载: 全尺寸图片幻灯片

图 4 Wine Quality数据集对CF-BLS与BLS算法的RMSE和PCC曲线

Fig. 4 Curves for RMSE and PCC of Wine Quality dataset based on CF-BLS and BLS

下载: 全尺寸图片幻灯片

图 2 Energy Efficiency数据集对CF-BLS与BLS算法的RMSE和PCC曲线

Fig. 2 Curves for RMSE and PCC of Energy Efficiency dataset based on CF-BLS and BLS

下载: 全尺寸图片幻灯片

图 3 Forest Fires数据集对CF-BLS与BLS算法的RMSE和PCC曲线

Fig. 3 Curves for RMSE and PCC of Forest Fires dataset based on CF-BLS and BLS

下载: 全尺寸图片幻灯片

图 5 两种模型的预测结果(100个增强节点)

Fig. 5 Predicted results of two model (100 enhancement nodes)

下载: 全尺寸图片幻灯片

图 6 两种模型的预测结果(100个特征节点)

Fig. 6 Predicted results of two model (100 feature nodes)

下载: 全尺寸图片幻灯片

图 7 基于CF-BLS, BLS, FBLS和HELM的虚拟机性能预测曲线

Fig. 7 Predicted curves for performance of virtual machine based on CF-BLS, BLS, FBLS, and HELM

下载: 全尺寸图片幻灯片

图 8 CF-BLS, BLS, FBLS和HELM模型的预测结果

Fig. 8 Predicted results of CF-BLS, BLS, FBLS, and HELM

下载: 全尺寸图片幻灯片

表 1 回归数据集

Table 1 Datasets of regression

回归数据集属性训练数据测试数据

Combined cycle power plant 4 4795 4773
Energy efficiency 8 468 300
Forest fires 13 259 258
Wine quality 12 2898 2000

下载: 导出CSV

参考文献(28)

[1]	李琳, 应时, 董波, 王蕊. 云环境中面向服务软件的演化部署优化方法. 中国科学: 信息科学, 2017, 47(6): 715-735 doi: 10.1360/N112016-00200 Li Lin, Ying Shi, Dong Bo, Wang Rui. Evolutionary deployment optimization for service-oriented software in cloud. <i>SCIENTIA SINICA Informationis</i>, 2017, 47(6): 715-735 doi: 10.1360/N112016-00200
[2]	罗军舟, 杨明, 凌振, 吴文甲, 顾晓丹. 网络空间安全体系与关键技术. 中国科学: 信息科学, 2016, 46(8): 939-968 doi: 10.1360/N112016-00090 Luo Jun-Zhou, Yang Ming, Ling Zhen, Wu Wen-Jia, Gu Xiao-Dan. Architecture and key technologies of cyberspace security. <i>SCIENTIA SINICA Informationis</i>, 2016, 46(8): 939-968 doi: 10.1360/N112016-00090
[3]	关兆雄, 庞维欣. 基于在线迁移的虚拟化资源整合研究. 自动化与仪器仪表, 2018, (3): 59-62 Guan Zhao-Xiong, Pang Wei-Xin. Research on virtual resource integration based on online migration. <i>Automation & Instrumentation</i>, 2018, (3): 59-62
[4]	余勇, 车建华, 徐焕良, 蒋诚智. 负载类型相关的Xen虚拟机系统性能模型. 计算机科学, 2016, 43(11): 210-214 doi: 10.11896/j.issn.1002-137X.2016.11.041 Yu Yong, Che Jian-Hua, Xu Huan-Liang, Jiang Cheng-Zhi. Workload type-dependent Xen virtual machine system performance models. <i>Computer Science</i>, 2016, 43(11): 210-214 doi: 10.11896/j.issn.1002-137X.2016.11.041
[5]	吕庆翰, 面向虚拟化的综合性能评测方法研究 [硕士学位论文], 华南理工大学, 中国, 2015. Lv Qing-Han. The Research about Virtualization Oriented Comprehensive Performance Evaluation Method [Master thesis], South China University of Technology, China, 2015.
[6]	车建华, 虚拟计算系统性能与可用性评测方法研究 [博士学位论文], 浙江大学, 中国, 2010. Che Jian-Hua. Research on Performance and Availability Evaluation Methods of Virtualization Systems [Ph.D. dissertation], Zhejiang University, China, 2010.
[7]	黎丰泽, 杨达, 周鹏, 武延军. 虚拟环境下虚拟机应用性能建模. 计算机系统应用, 2015, 24(9): 9-15 doi: 10.3969/j.issn.1003-3254.2015.09.002 Li Feng-Ze, Yang Da, Zhou Peng, Wu Yan-Jun. Modeling application performance in a virtualized environment. <i>Computer Systems & Applications</i>, 2015, 24(9): 9-15 doi: 10.3969/j.issn.1003-3254.2015.09.002
[8]	Xu J, Zhao M, Fortes J, Carpenter R, Yousif M. Autonomic resource management in virtualized data centers using fuzzy logic-based approaches. <i>Cluster Computing</i>, 2008, 11(3): 213-227 doi: 10.1007/s10586-008-0060-0
[9]	Rao J, Bu X P, Xu C Z, Wang L Y, Yin G. VCONF: A reinforcement learning approach to virtual machines auto-configuration. In: Proceedings of the 6th International Conference on Autonomic Computing. Barcelona, Spain: ACM, 2009. 137−146
[10]	贝振东, 喻之斌, 熊文, 林栋, 张慧玲, 须成忠. 一种云计算系统中虚拟机的性能预测方法及系统, 中国 104536829A, 2015-04-22 Bei Zhen-Dong, Yu Zhi-Bin, Xiong Wen, Lin Dong, Zhang Hui-Ling, Xu Cheng-Zhong. Performance Prediction Method and System for Virtual Machines in Cloud Computing System, CN Patent 104536829A, April 22, 2015
[11]	王娟, 张彬彬, 岳昆, 郝佳, 武浩. 基于随机森林回归的虚拟机性能预测方法, 中国 106897109A, 2017-06-27 Wang Juan, Zhang Bin-Bin, Yue Kun, Hao Jia, Wu Hao. Virtual Machine Performance Prediction Method Based on Random Forest Regression, CN Patent 106897109A, June 27, 2017
[12]	Chen C L P, Liu Z L. Broad learning system: An effective and efficient incremental learning system without the need for deep architecture. <i>IEEE Transactions on Neural Networks and Learning Systems</i>, 2018, 29(1): 10-24 doi: 10.1109/TNNLS.2017.2716952
[13]	郑云飞, 陈霸东. 基于最小p-范数的宽度学习系统. 模式识别与人工智能, 2019, 32(1): 51-57 Zheng Yun-Fei, Chen Ba-Dong. Least p-norm based broad learning system. <i>Pattern Recognition and Artificial Intelligence</i>, 2019, 32(1): 51-57
[14]	贾晨, 刘华平, 续欣莹, 孙富春. 基于宽度学习方法的多模态信息融合. 智能系统学报, 2019, 14(1): 150-157 Jia Chen, Liu Hua-Ping, Xu Xin-Ying, Sun Fu-Chun. Multi-modal information fusion based on broad learning method. <i>CAAI Transactions on Intelligent Systems</i>, 2019, 14(1): 150-157
[15]	Pao Y H, Park G H, Sobajic D J. Learning and generalization characteristics of the random vector functional-link net. <i>Neurocomputing</i>, 1994, 6(2): 163-180 doi: 10.1016/0925-2312(94)90053-1
[16]	管皓, 薛向阳, 安志勇. 深度学习在视频目标跟踪中的应用进展与展望. 自动化学报, 2016, 42(6): 834-847 Guan Hao, Xue Xiang-Yang, An Zhi-Yong. Advances on application of deep learning for video object tracking. <i>Acta Automatica Sinica</i>, 2016, 42(6): 834-847
[17]	刘巧元, 王玉茹, 张金玲, 殷明浩. 基于相关滤波器的视频跟踪方法研究进展. 自动化学报, 2019, 45(2): 265-275 Liu Qiao-Yuan, Wang Yu-Ru, Zhang Jin-Ling, Yin Ming-Hao. Research progress of visual tracking methods based on correlation filter. <i>Acta Automatica Sinica</i>, 2019, 45(2): 265-275
[18]	Feng S, Chen C L P. Fuzzy broad learning system: A novel neuro-fuzzy model for regression and classification. <i>IEEE Transactions on Cybernetics</i>, 2020, 50(2): 414-424 doi: 10.1109/TCYB.2018.2857815
[19]	Huang G B, Chen L, Siew C K. Universal approximation using incremental constructive feedforward networks with random hidden nodes. <i>IEEE Transactions on Neural Networks</i>, 2006, 17(4): 879-892 doi: 10.1109/TNN.2006.875977
[20]	张万栋, 李庆忠, 黎明, 武庆明. 基于最优误差自校正极限学习机的高频地波雷达RD谱图海面目标检测算法. 自动化学报, 2021, 47(1): 108-120 Zhang Wan-Dong, LI Qing-Zhong, LI Ming, Wu Qing-Ming. Sea surface target detection for RD images of HFSWR based on optimized error self-adjustment extreme learning machine. <i>Acta Automatica Sinica</i>, 2021, 47(1): 108-120
[21]	Tang J X, Deng C W, Huang G B. Extreme learning machine for multilayer perceptron. <i>IEEE Transactions on Neural Networks and Learning Systems</i>, 2016, 27(4): 809-821 doi: 10.1109/TNNLS.2015.2424995
[22]	焦李成, 赵进, 杨淑媛, 刘芳. 深度学习、优化与识别. 北京: 清华大学出版社, 2017. 123−125 Jiao Li-Cheng, Zhao Jin, Yang Shu-Yuan, Liu Fang. Deep Learning, Optimization and Recognition. Beijing: Tsinghua University Press, 2017. 123−125
[23]	Sabour S, Frosst N, Hinton G E. Dynamic routing between capsules. In: Proceedings of the 31st International Conference on Neural Information Processing Systems. Long Beach, USA: Curran Associates Inc. , 2017. 3859−3869
[24]	王博. 凸增量极限学习机的逼近阶 [硕士学位论文], 西北大学, 中国, 2015. Wang Bo. The Approximation Order of Convex Incremental Extreme Learning Machine [Master thesis], Northwest University, China, 2015.
[25]	田中大, 李树江, 王艳红, 王向东. 高斯过程回归补偿ARIMA的网络流量预测. 北京邮电大学学报, 2017, 40(6): 65-73 Tian Zhong-Da, Li Shu-Jiang, Wang Yan-Hong, Wang Xiang-Dong. Network traffic prediction based on ARIMA with Gaussian process regression compensation. <i>Journal of Beijing University of Posts and Telecommunications</i>, 2017, 40(6): 65-73
[26]	Tian Z D, Li S J, Wang Y H, Sha Y. A prediction method based on wavelet transform and multiple models fusion for chaotic time series. <i>Chaos, Solitons & Fractals</i>, 2017, 98: 158-172
[27]	Tian Z D, Ren Y, Wang G. Short-term wind speed prediction based on improved PSO algorithm optimized EM-ELM. <i>Energy Sources, Part A: Recovery, Utilization, and Environmental Effects</i>, 2019, 41(1): 26-46 doi: 10.1080/15567036.2018.1495782
[28]	UCI datasets [Online], available: http://archive.ics.uci.edu/ml/datasets.php, April 10, 2019