基于组合负载预测模型的多租户数据库弹性伸缩方法  

Elastic Scaling Method for Multi-tenant Databases Based on Hybrid Workload Prediction Model

作　　者：徐海洋 刘海龙[1,2] 陈先 王磊 金轲[1,2] 侯舒峰 李战怀 XU Hai-Yang;LIU Hai-Long;CHEN Xian;WANG Lei;JIN Ke;HOU Shu-Feng;LI Zhan-Huai(School of Computer science,Northwestern Polytechnical University,Xi’an 710100,China;Key Laboratory of Big Data Storage and Management(Northwestern Polytechnical University),Ministry of Industry and Information Technology,Xi’an 710100,China;Xi’an Zhongxing New Software Co.Ltd.,Xi’an 710100,China)

机构地区：[1]西北工业大学计算机学院,陕西西安710100 [2]大数据存储与管理工业和信息化部重点实验室(西北工业大学),陕西西安710100 [3]西安中兴新软件有限责任公司,陕西西安710100

出　　处：《软件学报》2025年第3期981-994,共14页Journal of Software

基　　金：国家重点研发计划(2023YFB4503600);国家自然科学基金(62172335);CCF-华为胡杨林基金(CCF-HuaweiDBIR0004B)。

摘　　要：云环境下的多租户数据库重要特性之一是可伸缩性,然而大部分的弹性伸缩技术难以针对复杂变化的负载进行有效的伸缩决策.若能提前预测负载变化,则能够准确地调整资源供给.鉴于此,提出了基于内存负载预测的多租户数据库弹性伸缩方法,包括一种组合负载预测模型和一种弹性伸缩策略.组合负载预测模型融合了卷积神经网络、长短期记忆网络和门控循环单元的优势,可以比较精确地预测数据库集群内存负载需求;弹性伸缩策略基于需求预测结果,调整虚拟机数目,保证资源供应处于合理范围.与现有方法对比,所提出的模型预测误差降低了8.7%-21.8%,预测拟合度提高了4.6%.在此基础上,改进了贝叶斯优化算法,用于该模型超参数调优,解决了贝叶斯优化在离散解、连续解的组合域中效果较差的问题,误差指标值降低了20%以上,拟合度提高了1.04%.实验结果表明,与Kubernetes中应用最广泛的伸缩策略相比,所提出的弹性伸缩方法避免了弹性伸缩的滞后性与资源浪费,响应时间降低了8.12%,延迟降低了9.56%.One of the most important features of multi-tenant databases in cloud environments is scalability.However,most elastic scaling techniques struggle to make effective scaling decisions for dynamically changing loads.If load changes can be predicted in advance,resource supply can be accurately adjusted.Given this,this study proposes a load-prediction-based elastic scaling method for multi-tenant databases.It includes a combined load prediction model and an elastic scaling strategy.The load prediction model combines the advantages of convolutional neural networks,long short-term memory networks and gated recurrent units.It can accurately forecast memory requirements of database clusters.Based on the prediction results,the elastic scaling strategy adjusts the number of virtual machines to ensure that resource supply remains within a reasonable range.Compared to existing methods,the combined load prediction model can reduce prediction errors by 8.7%to 21.8%and improve prediction fitting degree by 4.6%.Furthermore,this study improves the Bayesian optimization algorithm for hyperparameter tuning of the combined prediction model.The improved hyperparameter tuning model reduces errors by above 20%and improves fitting degree by 1.04%,which proves that it can well address the poor performance of Bayesian optimization in combined domains of discrete and continuous solutions.Compared to the most widely used scaling strategy in Kubernetes,the proposed elastic scaling method reduces response time by 8.12%and latency by 9.56%.It can avoid the latency and the waste of resources to a large extent.

关 键 词：多租户数据库 资源管理 资源预测 弹性伸缩 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]