Research on Optimization of Kubernetes Elastic Scaling Based on Entropy Weight Utilization and Prediction Algorithm

doi:10.19678/j.issn.1000-3428.0070139

Abstract

Abstract:

This study proposes an improved elastic scaling strategy based on a composite algorithm that combines entropy weight utilization and a prediction model, to address the issues of single-metric evaluation, latency, and low resource utilization in Kubernetes's built-in elastic scaling strategy. The entropy weight utilization composite algorithm calculates the comprehensive load value of the Kubernetes cluster by focusing on the distribution differences (information entropy method) and overall trends (average utilization weight method) of resource utilization across different nodes, thereby solving the problem of single metric evaluation. Next, this study constructs a predictive model that combines Adaptive Variational Mode Decomposition (AVMD) and the Attention Mechanism-based enhanced Long Short-Term Memory (Attention Mechanism-based LSTM) network to solve the latency and low resource utilization issues by predicting load changes. This model enables the system to quickly respond, expand its capacity at the onset of high traffic, and rapidly scale down to release resources once traffic subsides. Experimental results show that the improved elastic scaling strategy reduces the response time by 52% during the early stage of burst traffic compared with the default Kubernetes scaling strategy, and it rapidly scales down after the traffic subsides to release resources, demonstrating high practical application value.

Key words: Kubernetes cluster, entropy weight utilization composite algorithm, Adaptive Variational Mode Decomposition (AVMD) algorithm, Long Short-Term Memory (LSTM) algorithm, load prediction

摘要：

为解决Kubernetes内置的弹性伸缩策略衡量指标单一、反应滞后和资源利用效率低的问题, 提出一种熵权利用率复合算法结合预测模型的改进弹性伸缩策略。熵权利用率复合算法通过关注多种指标的资源利用率在不同节点上的分布差异(信息熵权法)和整体趋势(平均利用率权重法), 计算Kubernetes集群的综合负载值, 从而解决衡量指标单一的问题。构建自适应变分模态分解(AVMD)算法结合基于注意力机制增强的长短期记忆(Attention Mechanism-based LSTM)算法的预测模型, 通过预测负载变化以解决反应滞后和资源利用率低的问题。该模型根据预测的负载值, 在高流量初期促使系统快速响应进行扩容, 流量结束后迅速缩容以节约资源。实验结果表明, 与Kubernetes伸缩策略相比, 改进弹性伸缩策略在突发流量前期, 请求响应时间降低了52%, 在流量结束后快速缩容释放资源, 具有较高的实际应用价值。

关键词: Kubernetes集群, 熵权利用率复合算法, 自适应变分模态分解算法, 长短期记忆算法, 负载预测

SONG Zhedai, ZHU Jinrong, LIANG Chenyue, CHENG Xinyu. Research on Optimization of Kubernetes Elastic Scaling Based on Entropy Weight Utilization and Prediction Algorithm[J]. Computer Engineering, 2026, 52(4): 349-357.

宋哲代, 朱金荣, 梁琛悦, 程心雨. 基于熵权利用率与预测算法的Kubernetes弹性伸缩优化研究[J]. 计算机工程, 2026, 52(4): 349-357.

/ Recommend / Download Citations

URL: https://www.ecice06.com/EN/10.19678/j.issn.1000-3428.0070139

https://www.ecice06.com/EN/Y2026/V52/I4/349

Figures/Tables 9

Fig.1 The LSTM working procedure of AVMD combined attention mechanism

Fig.2 Comparison between AVMD combined with Attention Mechanism LSTM model and real data

Fig.3 Pod number change

Fig.4 Response time change

Fig.5 Pod number change

Fig.6 Response time change

References 26

1	ERMOLENKO D, KILICHEVA C, MUTHANNA A, et al. Internet of things services orchestration framework based on Kubernetes and edge computing[C]//Proceeding of the IEEE Conference of Russian Young Researchers in Electrical and Electronic Engineering. Washington D. C., USA: IEEE Press, 2021: 12-17.
2	DING Z , WANG S , JIANG C . Kubernetes-oriented microservice placement with dynamic resource allocation. IEEE Transactions on Cloud Computing, 2023, 11, 1777- 1793. doi: 10.1109/TCC.2022.3161900
3	李佳颖, 杨泽民, 宋哲代, 等. Kubernetes容器云的弹性伸缩方法研究. 电子科技, 2025 (3): 52- 59.
	LI J Y , YANG Z M , SONG Z D . Research on elastic scaling method of Kubernetes container cloud. Electronic Science and Technology, 2025 (3): 52- 59.
4	ROSSI F , CARDELLINI V , PRESTI F L , et al. Dynamic multi-metric thresholds for scaling applications using reinforcement learning. IEEE Transactions on Cloud Computing, 2023, 11 (2): 1807- 1821. doi: 10.1109/TCC.2022.3163357
5	沐磊, 李洪赭, 李赛飞. 一种改进的Kubernetes弹性伸缩策略. 计算机与数字工程, 2022, 50 (2): 327-331, 372.
	MU L , LI H Z , LI S F . An improved elastic scaling strategy for Kubernetes. Computer and Digital Engineering, 2022, 50 (2): 327-331, 372.
6	夏冰冰, 范中磊. 一种预测与响应相结合的Kubernetes容器云弹性伸缩策略. 微电子学与计算机, 2025, 42 (3): 40- 48.
	XIA B B , FAN Z L . A prediction and response combined elastic scaling strategy for Kubernetes container cloud. Microelectronics & Computer, 2025, 42 (3): 40- 48.
7	LIU Y B L , ZHANG Y , YU S , et al. An efficient new adaptive variational mode decomposition algorithm for extracting adventitious lung sounds. Biomedical Signal Processing and Control, 2024, 89, 342- 351.
8	JAIN D K , JAIN R , UPADHYAY Y , et al. Deep refinement: capsule network with attention mechanism-based system for text classification. Neural Computing and Applications, 2020, 32 (7): 1839- 18. doi: 10.1007/s00521-019-04620-z
9	ZHU Y X , TIAN D Z , FENG Y . Effectiveness of entropy weight method in decision-making. Mathematical Problems in Engineering, 2020, 20 (1): 224- 236.
10	PATEL D K , TRIPATHY D , TRIPATHY C R . Survey of load balancing techniques for grid. Journal of Network and Computer Applications, 2016, 65, 103- 119. doi: 10.1016/j.jnca.2016.02.012
11	ROSSI F, CARDELLINI V, PRESTI F L. Self-adaptive threshold-based policy for microservices elasticity[C]//Proceedings of the 28th International Symposium on Modeling, Analysis, and Simulation of Computer and Telecommunication Systems. Washington D. C., USA: IEEE Press, 2020: 331-342.
12	SHAFIQ D A , JHANJHI N Z , ABDULLAH A . Load balancing techniques in cloud computing environment: a review. Journal of King Saud University-Computer and Information Sciences, 2022, 34 (7): 3910- 3933. doi: 10.1016/j.jksuci.2021.02.007
13	IMDOUKH M , AHMAD I , ALFAILAKAWI M G . Machine learning-based auto-scaling for containerized applications. Neural Computing and Applications, 2020, 32 (13): 9745- 9760. doi: 10.1007/s00521-019-04507-z
14	蔡亮, 鲁家南, 才振功, 等. 一种基于历史数据分析的容器云平台资源配额预测方法[P]. 中国专利: CN201911360632.8, 2023-05-23.
	CAI L, LU J N, CAI Z G, et al. A resource quota prediction method for container cloud platform based on historical data analysis[P]. Chinese patent: CN201911360632.8, 2023-05-23. (in Chinese)
15	SENJAB K , ABBAS S , AHMED N , et al. A survey of Kubernetes scheduling algorithms. Journal of Cloud Computing, 2023, 12 (1): 87. doi: 10.1186/s13677-023-00471-1
16	ZHONG W , ZHUANG Y , SUN J , et al. A load prediction model for cloud computing using PSO-based weighted wavelet support vector machine. Applied Intelligence, 2018, 48 (11): 4072- 4083. doi: 10.1007/s10489-018-1194-2
17	SU C T , SHIUE Y R . Intelligent scheduling controller for shop floor control systems: a hybrid genetic algorithm/decision tree learning approach. International Journal of Production Research, 2003, 41 (12): 2619- 2641. doi: 10.1080/0020754031000090612
18	WANG L , CHE L , LAM Y K , et al. Mobile traffic prediction with attention-based hybrid deep learning. Physical Communication, 2024, 66, 102420- 102420. doi: 10.1016/j.phycom.2024.102420
19	胡国乐, 李鹏, 林事力, 等. 基于相位变换和CNN-BiLSTM的自动调制识别算法. 电讯技术, 2024, 64 (11): 1780- 1787.
	HU G L , LI P , LIN S L , et al. An automatic modulation recognition algorithm based on phase transformation and CNN-BiLSTM. Telecommunication Engineering, 2024, 64 (11): 1780- 1787.
20	LI M X , YAN C , LIU W , et al. Fault diagnosis model of rolling bearing based on parameter adaptive AVMD algorithm. Applied Intelligence, 2023, 53 (3): 3150- 3165. doi: 10.1007/s10489-022-03562-9
21	GU R , CHEN J , HONG R , et al. Incipient fault diagnosis of rolling bearings based on adaptive variational mode decomposition and Teager energy operator. Measurement, 2020, 149, 106941. doi: 10.1016/j.measurement.2019.106941
22	ZHOU J , WANG J . A novel underdetermined source number estimation for coupled vibration sources of mechanical fault based on variational mode decomposition. Journal of Mechanical Science and Technology, 2022, 36, 621- 635. doi: 10.1007/s12206-022-0110-1
23	游卉擎, 黄鹏程, 赵振宇, 等. 基于RNN的标准单元延时预测方法. 郑州大学学报(理学版), 2025, 57 (3): 28- 34.
	YOU H Q , HUANG P C , ZHAO Z Y , et al. Delay prediction method of standard cell based on RNN. Journal of Zhengzhou University (Natural Science Edition), 2025, 57 (3): 28- 34.
24	NG H R , ZHONG X , NAM Y , et al. Deep-learning-based approach for automated detection of irregular walking surfaces for walkability assessment with wearable sensor. Applied Sciences, 2023, 13 (24): 13053. doi: 10.3390/app132413053
25	PENG T M , HUBEELE N F , KARADY G G . Advancement in the application of neural networks for short-term load forecasting. IEEE Transactions on Power Systems, 1992, 45 (7): 250- 257.
26	李俊俊, 董建刚, 李坤. 基于Kubernetes的集群节能策略研究. 计算机工程, 2024, 50 (9): 82- 91. doi: 10.19678/j.issn.1000-3428.0068453
	LI J J , DONG J G , LI K . Research on Kubernetes-based cluster energy-saving strategy. Computer Engineering, 2024, 50 (9): 82- 91. doi: 10.19678/j.issn.1000-3428.0068453

[1]	CHEN Hao, DANG Zheng, HEI Xinhong, ZHAO Tong, ZHANG Jie. Load Prediction of Digital Twin Logical Range in Space-Air-Ground Computing Power Networks Based on SSA-VMD [J]. Computer Engineering, 2025, 51(5): 20-32.
[2]	ZHU Li, XIA Yu, ZHU Chunqiang, DENG Fan. Power Load Prediction Method Based on CEEMDAN and Spectral Time Graph Convolutional Networks [J]. Computer Engineering, 2025, 51(4): 339-349.
[3]	Haoyang LI, Xiaowei HE, Bin WANG, Hao WU, Qi YOU. Cloud Computing Resource Load Prediction Based on Improved Informer [J]. Computer Engineering, 2024, 50(2): 43-50.
[4]	Shiming PENG, Shiyang LIN, Shuo JIA, Miaohui YANG. Multi-Objective Optimization Task Offloading Strategy Based on Load Prediction [J]. Computer Engineering, 2024, 50(1): 206-215.
[5]	YANG Kaiqi, YAO Pei, ZHAO Yulong, TANG Lingtao. Application Migration Method for Heterogeneous Container Cloud [J]. Computer Engineering, 2019, 45(8): 42-47.
[6]	YAO Jiaxin,TIAN Huixin. Power Load Prediction Based on Improved Data-driven Subspace Algorithm [J]. Computer Engineering, 2015, 41(5): 311-315.
[7]	XU Jian. Research and Application on Load Model of Billing System [J]. Computer Engineering, 2013, 39(6): 82-84,90.
[8]	ZHANG Shu-Ben, XU Chen-Feng, XI Hong-Sheng. Load Balancing Algorithm for Middleware Database Cluster [J]. Computer Engineering, 2012, 38(5): 41-43,46.
[9]	WEN Shao-Jun, CHEN Dun-Jie, GUO Chao. Optimized Virtual Machine Deployment Mechanism in Cloud Platform [J]. Computer Engineering, 2012, 38(11): 17-19.
[10]	MA Wei-Jun, FENG Jing, CHEN Ye, XU Si-Lin. Fault Detection Adaptive Algorithm in Distributed Storage System [J]. Computer Engineering, 2012, 38(10): 257-259.
[11]	HAN Dui, GU Zhen-Gong, QIN Ti-Zhong, CHANG Chun, WANG Gao. Telephone Traffic Load Prediction Based on SVR with DE-strategy [J]. Computer Engineering, 2011, 37(2): 178-179.
[12]	YANG Wei; ZHU Qiaoming; LI Peifeng; QIAN Peide;. Server Load Prediction Based on Time Series [J]. Computer Engineering, 2006, 32(19): 143-145,.

Please choose a citation manager

Content to export