计算机科学

计算机科学 ›› 2024, Vol. 51 ›› Issue (2): 27-35.doi: 10.11896/jsjkx.231000141

• 数据库&大数据&数据科学 • 上一篇    下一篇

MMOS:支持超卖的多租户数据库内存资源共享方法

徐海洋1,2, 刘海龙1,2, 杨超云1, 王硕1,2, 李战怀1,2   

  1. 1 西北工业大学计算机学院 西安710100
    2 西北工业大学大数据存储与管理工业和信息化部重点实验室 西安710100
  • 收稿日期:2023-10-20 修回日期:2023-11-24 出版日期:2024-02-15 发布日期:2024-02-22
  • 通讯作者: 刘海龙(liuhailong@nwpu.edu.cn)
  • 作者简介:(1299895191@qq.com)
  • 基金资助:
    国家自然科学基金(62172335);CCF-华为胡杨林基金(CCF-HuaweiDBIR0004B)

MMOS:Memory Resource Sharing Methods to Support Overselling in Multi-tenant Databases

XU Haiyang1,2, LIU Hailong1,2, YANG Chaoyun1, WANG Shuo1,2, LI Zhanhuai1,2   

  1. 1 School of Computer Science,Northwestern Polytechnical University,Xi'an 710100,China
    2 Key Laboratory of Big Data Storage and Management,Northwestern Polytechnical University,Ministry of Industry and Information Techno- logy,Xi'an 710100,China
  • Received:2023-10-20 Revised:2023-11-24 Online:2024-02-15 Published:2024-02-22
  • About author:XU Haiyang,born in 2000,postgra-duate,is a member of CCF(No.H3613G).His main research interests include big data management and analysis technology.LIU Hailong,born in 1980,Ph.D,associate professor,is a member of CCF(No.10265M).His main research interests include big data management and analysis technologies,data gover-nance technologies.
  • Supported by:
    National Natural Science Foundation of China (62172335) and CCF-Huawei Hoopwood Fund(CCF-HuaweiDBIR0004B).

PDF (PC)

1066

摘要： 多租户数据库为每个租户分配固定的资源配额,而这些资源配额通常未全部得到有效利用,这种静态分配策略导致资源利用率不高。若在不影响租户性能的前提下将未利用的空闲资源共享给其他租户使用,即实现资源超卖,则可以提高资源利用率、提升平台收益。为了支持资源超卖,需要准确预测租户的资源需求,动态地按需为租户分配资源。已有的针对多租户数据库的资源共享方法的研究对象主要是CPU资源,鲜有支持超卖的内存资源共享方法。鉴于此,在联机分析处理场景下,提出了一种支持超卖的多租户数据库内存资源共享方法MMOS(Multi-tenant database Memory resource Overselling and Sharing)。该方法通过准确预测每个租户的内存需求区间,按照区间上限为租户动态调整内存配额,在不影响租户性能的前提下,统一管理空闲内存资源以支持更多租户,实现内存超卖。实验结果表明,MMOS在租户负载动态变化的场景下具有较好效果。在不同资源量的资源池下,支持的租户数可以增加2~2.6倍,资源利用率峰值提升175%~238%。同时,每个租户的业务与性能未受影响。

关键词: 多租户数据库, 资源超卖, 内存资源, 资源预测, 资源分配

Abstract: This paper presents an oversold memory resource sharing method for multi-tenant databases in an online analysis and processing scenario.The current static resource allocation strategy,which assigns a fixed resource quota to each tenant,leads to suboptimal resource utilization.To enhance resource utilization and platform revenue,it is important to share unused free resources among tenants without impacting their performance.While existing resource sharing methods for multi-tenant databases primarily focus on CPU resources,there is a lack of memory resource sharing methods that support overselling.To address this gap,the paper introduces a novel approach MMOS that accurately forecasts the memory requirements interval of each tenant and dynamically adjusts their resource allocation based on the upper limit of the interval.This allows for efficient management of free memory resources,enabling support for more tenants and achieving memory overselling while maintaining optimal performance.Experimental results demonstrate the effectiveness of the proposed method in dynamically changing tenant load scenarios.With different resource pools,the number of supported tenants can be increased by 2~2.6 times,leading to a significant increase in peak resource utilization by 175%~238%.Importantly,the proposed method ensures that the business and performance of each tenant remain unaffected.

Key words: Multi-tenant database, Resource overselling, Memory resources, Resource forecasting, Resource allocation

中图分类号: 

  • TP311.13
[1]RASHID A,CHATURVEDI A.Cloud computing characteris-tics and services:a brief review[J].International Journal of Computer Sciences and Engineering,2019,7(2):421-426.
[2]NARASAYYA V,CHAUDHURI S.Multi-Tenant Cloud Data Services:State-of-the-Art,Challenges and Opportunities[C]//Proceedings of the 2022 ACM SIGMOD International Confe-rence on Management of Data.2022:2465-2473.
[3]LIN W,XU S,HE L,et al.Multi-resource scheduling and power simulation for cloud computing[J].Information Sciences—Informatics and Computer Science,Intelligent Systems,Applications:An International Journal,2017,397(C):168-186.
[4]KUMAR K S S,JAISANKAR N.An automated resource mana-gement framework for minimizing SLA violations and negotiation in collaborative cloud[J].International Journal of Cognitive Computing in Engineering,2020,1:27-35.
[5]THEIN T,MYO M M,PARVIN S,et al.Reinforcement lear-ning based methodology for energy-efficient resource allocation in cloud data centers[J].Journal of King Saud University-Computer and Information Sciences,2020,32(10):1127-1139.
[6]ZHAO Y,CALHEIROS R N,GANGE G,et al.SLA-basedprofit optimization resource scheduling for big data analytics-as-a-service platforms in cloud computing environments[J].IEEE Transactions on Cloud Computing,2018,9(3):1236-1253.
[7]WANG Y,HE Q,ZHANG X,et al.Efficient QoS-aware service recommendation for multi-tenant service-based systems in cloud[J].IEEE Transactions on Services Computing,2017,13(6):1045-1058.
[8]LIU L.Qos-aware machine learning-based multiple resourcesscheduling for microservices in cloud environment[J].arXiv:1911.13208,2019.
[9]QI B,ZHANG P,WU H,et al.Cloud Resource SchedulingMethod based on Markov Process and the Cuckoo Search[C]//Journal of Physics:Conference Series.IOP Publishing,2022,2320(1):12-30.
[10]IMDOUKH M,AHMAD I,ALFAILAKAWI M G.Machinelearning-based auto-scaling for containerized applications[J].Neural Computing and Applications,2020,32:9745-9760.
[11]SINGH S T,TIWARI M,DHAR A S.Machine Learning based Workload Prediction for Auto-scaling Cloud Applications[C]//2022 OPJU International Technology Conference on Emerging Technologies for Sustainable Development(OTCON).IEEE,2023:1-6.
[12]HIGGINSON A S,DEDIU M,ARSENE O,et al.Databaseworkload capacity planning using time series analysis and machine learning[C]//Proceedings of the 2020 ACM SIGMOD,International Conference on Management of Data.2020:769-783.
[13]KHOLIDY H A.An intelligent swarm based prediction ap-proach for predicting cloud computing user resource needs[J].Computer Communications,2020,151:133-144.
[14]CHEN J,WANG Y.A Hybrid Method for Short-Term HostUtilization Prediction in Cloud Computing[J].Journal of Electrical & Computer Engineering,2019,2019:33-46.
[15]QIU C,SHEN H.Dynamic demand prediction and allocation in cloud service brokerage[J].IEEE Transactions on Cloud Computing,2019,9(4):1439-1452.
[16]QADER W A,AMEEN M M,AHMED B I.An overview of bag of words;importance,implementation,applications,and challenges[C]//2019 International Engineering Conference(IEC).IEEE,2019:200-204.
[17]LI Y,YANG T.Word embedding for understanding natural language:a survey[J].Guide to Big Data Applications,2018,4(4):83-104.
[18]ZHAO C,WANG S,FENG X,et al.An Improved Term Frequency-Inverse Document Frequency Method Solving Multi-Text Label Problem[C]//2022 Global Conference on Robotics,Artificial Intelligence and Information Technology(GCRAIT).IEEE,2022:400-404.
[19]CHEN T,GUESTRIN C.Xgboost:A scalable tree boostingsystem[C]//Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.2016:785-794.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市渝北区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发