计算机科学 ›› 2021, Vol. 48 ›› Issue (4): 261-267.doi: 10.11896/jsjkx.200800023

• 计算机网络 • 上一篇    下一篇

无服务器平台资源调度综述

马泽华1, 刘波1, 林伟伟2, 李加伟1   

  1. 1 华南师范大学计算机学院 广州510631
    2 华南理工大学计算机科学与工程学院 广州510641
  • 收稿日期:2020-06-24 修回日期:2020-09-10 出版日期:2021-04-15 发布日期:2021-04-09
  • 通讯作者: 林伟伟(linww@scut.edu.cn)
  • 基金资助:
    国家自然科学基金项目(61872084,61772205);广东省基础与应用基础研究重大项目(2019B030302002);广州市科技计划项目 (202007040002,201902010040)

Survey of Resource Scheduling for Serverless Platforms

MA Ze-hua1, LIU Bo1, LIN Wei-wei2, LI Jia-wei1   

  1. 1 School of Computer Science,South China Normal University,Guangzhou 510631,China
    2 School of Computer Science and Engineering,South China University of Technology,Guangzhou 510641,China
  • Received:2020-06-24 Revised:2020-09-10 Online:2021-04-15 Published:2021-04-09
  • About author:MA Ze-hua,born in 1996,postgraduate.His main research interests include cloud computing and container scheduling.(mzh.scnu@qq.com)
    LIN Wei-wei,born in 1980,Ph.D,professor,is a member of China Computer Federation.His main research interests include cloud computing,big data technology and AI application technology.
  • Supported by:
    National Natural Science Foundation of China (61872084,61772205),Guangdong Major Project of Basic and Applied Basic Research (2019B030302002) and Guangzhou Science and Technology Plan Project (202007040002,201902010040).

摘要: 无服务器(Serverless)计算正成为部署云应用程序的一种具有广阔发展前景的范式。其实现了一种真正的现收现付的计费方式,并且不会浪费资源。开发人员无需担心计算平台的底层细节,只需在处理请求或事件时付费,从而降低了开发人员的门槛。无服务器模式的转变虽然带来了机遇,但也带来了平台函数冷启动延迟、资源利用不足等问题。为此,文中对无服务器计算平台的资源调度技术做了深入的调查和分析,重点阐述了面向资源利用、响应时间延迟以及多目标优化的无服务器平台资源调度的技术原理和相关研究现状,并在此基础上分析总结并指明了无服务器平台资源调度未来的主要研究方向:即面向不同应用类型负载的调度优化、响应时间与资源利用率的折衷调度、虚拟机与无服务器平台的联合调度以及无服务器资源调度的混合算法。

关键词: 函数即服务, 无服务器计算, 资源调度

Abstract: Serverless computing is emerging as a promising paradigm for deploying cloud applications.It really implements pay-as-you-go billing without wasting resources.Developers don’t have to worry about the low-level details of the computing platform,they just need to pay when processing requests or events,thus lowering the threshold for developers.Although the change of serverless model brings opportunities,it also brings problems such as cold startup delay of platform function and insufficient resource utilization.Therefore,this paper makes an in-depth investigation and analysis of the resource scheduling technology of the serverless computing platform.The technical principle and research status of resource scheduling in serverless platform based on resource utilization,response time delay and multi-objective optimization are discussed.Finally,this paper points out the trends of future research directions:scheduling for different application types,the tradeoff between response time and resource utilization,joint scheduling between virtual machine and serverless platform,and hybrid algorithm for serverless resource scheduling.

Key words: Functions as a services, Resource scheduling, Serverless computing

中图分类号: 

  • TP393
[1]KILCIOGLU C,RAO J M,KANNAN A,et al.Usage patterns and the economics of the public cloud[C]//Proceedings of the 26th International Conference on World Wide Web.2017:83-91.
[2]CNCF Serverless Whitepaper v1.0[EB/OL].(2020-05-23)[2020-08-31].https://github.com/cncf/wg-serverless#white-paper.
[3]MCGRATH G,BRENNER P R.Serverless computing:Design,implementation,and performance[C]//2017 IEEE 37th International Conference on Distributed Computing Systems Workshops(ICDCSW).IEEE,2017:405-410.
[4]Global Forecast to 2021:Increasing shift from DevOps to serverless computing to drive the overall Function-as-a-Service market[EB/OL].(2017-06-15)[2020-08-31].https://www.businesswire.com/news/home/20170227006262/en/7.72-Billion-Function-as-a-Service-Market-2017---Global.
[5]DELIMITROU C,KOZYRAKIS C.Quasar:resource-efficientand QoS-aware cluster management[J].ACM SIGPLAN Notices,2014,49(4):127-144.
[6]GAN Y,ZHANG Y,CHENG D,et al.An open-source benchmark suite for microservices and their hardware-software implications for cloud & edge systems[C]//Proceedings of the Twenty-Fourth International Conference on Architectural Support for Programming Languages and Operating Systems.2019:3-18.
[7]MAHMOUDI N,LIN C,KHAZAEI H,et al.Optimizing serverless computing:introducing an adaptive function placement algorithm[C]//Proceedings of the 29th Annual International Conference on Computer Science and Software Engineering.2019:203-213.
[8]CASTRO P,ISHAKIAN V,MUTHUSAMY V,et al.The rise of serverless computing[J].Communications of the ACM,2019,62(12):44-54.
[9]ISHAKIAN V,MUTHUSAMY V,SLOMINSKI A.Servingdeep learning models in a serverless platform[C]//2018 IEEE International Conference on Cloud Engineering(IC2E).IEEE,2018:257-262.
[10]BALDINI I,CASTRO P,CHANG K,et al.Serverless computing:Current trends and open problems[M]//Research Advances in Cloud Computing.Springer,Singapore,2017:1-20.
[11]HELLERSTEIN J M,FALEIRO J,GONZALEZ J E,et al.Serverless computing:One step forward,two steps back[J].arXiv:1812.03651,2018.
[12]FOX G C,ISHAKIAN V,MUTHUSAMY V,et al.Status of serverless computing and function-as-a-service(faas) in industry and research[J].arXiv:1708.08028,2017.
[13]ADZIC G,CHATLEY R.Serverless computing:economic andarchitectural impact[C]//Proceedings of the 2017 11th Joint Meeting on Foundations of Software Engineering.2017:884-889.
[14]JONAS E,SCHLEIER-SMITH J,SREEKANTI V,et al.Cloud programming simplified:A berkeley view on serverless computing[J].arXiv:1902.03383,2019.
[15]WANG L,LI M,ZHANG Y,et al.Peeking behind the curtains of serverless platforms[C]//Annual Technical Conference.2018:133-146.
[16]SURESH A,GANDHI A.FnSched:An Efficient Scheduler for Serverless Functions[C]//Proceedings of the 5th International Workshop on Serverless Computing.2019:19-24.
[17]LEE H,SATYAM K,FOX G.Evaluation of production serverless computing environments[C]//2018 IEEE 11th Internatio-nal Conference on Cloud Computing(CLOUD).IEEE,2018:442-450.
[18]WAGNER T.Optimizing Enterprise Economics with Serverless Architectures[Z].AWS Web Serviced,2017.
[19]Netflix & AWS Lambda Case Study[EB/OL].(2020-08-18)[2020-08-31].https://aws.amazon.com/cn/solutions/case-studies/netflix-and-aws-lambda/.
[20]WU D,ROSEN D W,WANG L,et al.Cloud-based design and manufacturing:A new paradigm in digital manufacturing and design innovation[J].Computer-Aided Design,2015,59:1-14.
[21]KAFFES K,YADWADKAR N J,KOZYRAKIS C.Centralized Core-granular Scheduling for Serverless Functions[C]//Proceedings of the ACM Symposium on Cloud Computing.2019:158-164.
[22]JIANG Q,LEE Y C,ZOMAYA A Y.Serverless execution of scientific workflows[C]//International Conference on Service-Oriented Computing.Springer,Cham,2017:706-721.
[23]PÉREZ A,MOLTÓ G,CABALLER M,et al.Serverless compcuting for container-based architectures[J].Future Generation Computer Systems,2018,83:50-59.
[24]BALDINI I,CHENG P,FINK S J,et al.The serverless trilemma:Function composition for serverless computing[C]//Proceedings of the 2017 ACM SIGPLAN International Symposium on New Ideas,New Paradigms,and Reflections on Programming and Software.2017:89-103.
[25]LLOYD W,VU M,ZHANG B,et al.Improving Application Migration to Serverless Computing Platforms:Latency Mitigation with Keep-Alive Workloads[C]//2018 IEEE/ACM Internatio-nal Conference on Utility and Cloud Computing Companion(UCC Companion).IEEE,2018:195-200.
[26]MANNER J,ENDREβ M,HECKEL T,et al.Cold start influencing factors in function as a service[C]//2018 IEEE/ACM International Conference on Utility and Cloud Computing Companion(UCC Companion).IEEE,2018:181-188.
[27]VAN E E,IOSUP A,ABAD C L,et al.A SPEC RG cloudgroup’s vision on the performance challenges of FaaS cloud architectures[C]//Companion of the 2018 ACM/SPEC International Conference on Performance Engineering.2018:21-24.
[28]ABAD C L,BOZA E F,VAN EYK E.Package-aware scheduling of FaaS functions[C]//Companion of the 2018 ACM/SPEC International Conference on Performance Engineering.2018:101-106.
[29]OAKES E,YANG L,HOUCK K,et al.Pipsqueak:Lean Lambdas with large libraries[C]//2017 IEEE 37th International Conference on Distributed Computing Systems Workshops(ICDCSW).IEEE,2017:395-400.
[30]TOTOY G,BOZA E F,ABAD C L.An Extensible Scheduler for the OpenLambda FaaS Platform[C]//Min-Move workshop(co-located with ASPLOS).2018.
[31]HENDRICKSON S,STURDEVANT S,HARTER T,et al.Serverless computation with openlambda[C]//8th(USENIX)Workshop on Hot Topics in Cloud Computing(HotCloud 16).2016.
[32]KARGER D,SHERMAN A,BERKHEIMER A,et al.Web caching with consistent hashing[J].Computer Networks,1999,31(11-16):1203-1213.
[33]STOICA I,MORRIS R,KARGER D,et al.Chord:A scalablepeer-to-peer lookup service for internet applications[J].ACM SIGCOMM Computer Communication Review,2001,31(4):149-160.
[34]NASIR A,MORALES G D F,GARCíA-SORIANO D,et al.The Power of Both Choices:Practical Load Balancing for Distributed Stream Processing Engines[J].arXiv:1504.00788v1,2015.
[35]AUMALA G,BOZA E,ORTIZ-AVILES L,et al.Beyond Load Balancing:Package-Aware Scheduling for Serverless Platforms[C]//2019 19th IEEE/ACM International Symposium on Cluster,Cloud and Grid Computing(CCGRID).2019:282-291.
[36]BERMBACH D,KARAKAYA A S,BUCHHOLZ S.Using application knowledge to reduce cold starts in FaaS services[C]//Proceedings of the 35th Annual ACM Symposium on Applied Computing.2020:134-143.
[37]RODRIGUEZ M A,BUYYA R.Deadline based resource provisioningand scheduling algorithm for scientific workflows on clouds[J].IEEE transactions on cloud computing,2014,2(2):222-235.
[38]RYAN T,LEE Y C.Effective resource multiplexing for scienti-fic workflows[C]//2015 17th Asia-Pacific Network Operations and Management Symposium(APNOMS).IEEE,2015:232-237.
[39]ALQARYOUTI O,SIYAM N.Serverless Computing andScheduling Tasks on Cloud:A Review[J].American Scientific Research Journal for Engineering,Technology,and Sciences(ASRJETS),2018,40(1):235-247.
[40]STEIN M.The Serverless Scheduling Problem and NOAH[J].arXiv:1809.06100v1,2018.
[41]QUEVEDO S,MERCHÁN F,RIVADENEIRA R,et al.Evaluating Apache OpenWhisk-FaaS[C]//2019 IEEE Fourth Ecuador Technical Chapters Meeting(ETCM).IEEE,2019:1-5.
[42]HOSEINYFARAHABADY M R,LEE Y C,ZOMAYA A Y,et al.A QoS-aware resource allocation controller for function as a service(FaaS) platform[C]//International Conference on Ser-vice-Oriented Computing.Springer,Cham,2017:241-255.
[43]ENES J,EXPÓSITO R R,TOURIÑO J.Real-time resourcescaling platform for Big Data workloads on serverless environments[J].Future Generation Computer Systems,2020,105:361-379.
[44]GAND F,FRONZA I,EL IOINI N,et al.Serverless Container Cluster Management for Lightweight Edge Clouds[C]//CLOSER.2020:302-311.
[45]NASTIC S,RAUSCH T,SCEKIC O,et al.A serverless real-time data analytics platform for edge computing[J].IEEE Internet Computing,2017,21(4):64-71.
[46]CICCONETTI C,CONTI M,PASSARELLA A.An architectu-ral framework for serverless edge computing:design and emulation tools[C]//2018 IEEE International Conference on Cloud Computing Technology and Science(CloudCom).IEEE,2018:48-55.
[47]ZHANG H,STAFMAN L,OR A,et al.Slaq:quality-drivenscheduling for distributed machine learning[C]//Proceedings of the 2017 Symposium on Cloud Computing.2017:390-404.
[48]MARINI F,WALCZAK B.Particle swarm optimization(PSO).A tutorial[J].Chemometrics and Intelligent Laboratory Systems,2015,149:153-165.
[49]GUNTSCH M,MIDDENDORF M.A population based ap-proach for ACO[C]//Workshops on applications of evolutionary computation.Berlin,Heidelberg:Springer,2002:72-81.
[50]WANG J,CHEN H.BSAS:Beetle swarm antennae search algorithm for optimization problems[J].arXiv:1807.10470,2018.
[1] 柳鹏, 刘波, 周娜琴, 彭心怡, 林伟伟.
混合云工作流调度综述
Survey of Hybrid Cloud Workflow Scheduling
计算机科学, 2022, 49(5): 235-243. https://doi.org/10.11896/jsjkx.210300303
[2] 宋海宁, 焦健, 刘永.
高速公路中的移动边缘计算研究
Research on Mobile Edge Computing in Expressway
计算机科学, 2021, 48(6A): 383-386. https://doi.org/10.11896/jsjkx.200900212
[3] 宁玉辉, 姚喜.
一种应急指挥系统的设计与实现
Design and Implementation of Emergency Command System
计算机科学, 2021, 48(6A): 613-618. https://doi.org/10.11896/jsjkx.201000136
[4] 简琤峰, 平靖, 张美玉.
面向边缘计算的Storm边缘节点调度优化方法
Edge Computing-oriented Storm Edge Node Scheduling Optimization Method
计算机科学, 2020, 47(5): 277-283. https://doi.org/10.11896/jsjkx.190600048
[5] 徐飞, 王少昌, 杨卫霞.
基于博弈论的云资源调度算法
Cloud Resource Scheduling Algorithm Based on Game Theory
计算机科学, 2019, 46(6A): 295-299.
[6] 汪晨欣, 杨家海, 庄奕, 罗念龙.
未来网络试验设施的节点资源调度算法
Node Resource Scheduling for Future Network Experimentation Facility
计算机科学, 2019, 46(12): 95-100. https://doi.org/10.11896/jsjkx.190400106
[7] 赵宏伟, 田力威.
基于改进细菌觅食算法的云计算资源调度策略
Cloud Computing Resource Scheduling Strategy Based on Improved Bacterial Foraging Algorithm
计算机科学, 2019, 46(11): 309-314. https://doi.org/10.11896/jsjkx.181002000
[8] 阳小兰,钱程,朱福喜.
基于云计算的大数据服务资源评价方法
Evaluation Method of Big Data Service Resources Based on Cloud Computing
计算机科学, 2018, 45(5): 295-299. https://doi.org/10.11896/j.issn.1002-137X.2018.05.051
[9] 赵宏伟,李圣普.
基于粒子群算法和RBF神经网络的云计算资源调度方法研究
Research on Resources Scheduling Method in Cloud Computing Based on PSO and RBF Neural Network
计算机科学, 2016, 43(3): 113-117. https://doi.org/10.11896/j.issn.1002-137X.2016.03.023
[10] 张恒巍,韩继红,卫波,王晋东.
基于Map-Reduce模型的云资源调度方法研究
Research on Cloud Resource Scheduling Method Based on Map-Reduce
计算机科学, 2015, 42(8): 118-123.
[11] 周丽娟,王春影.
基于粒子群优化算法的云计算资源调度策略研究
Cloud Computing Resource Scheduling in Mobile Internet Based on Particle Swarm Optimization Algorithm
计算机科学, 2015, 42(6): 279-281. https://doi.org/10.11896/j.issn.1002-137X.2015.06.058
[12] 袁 浩,李昌兵.
基于社会力群智能优化算法的云计算资源调度
Resource Scheduling Algorithm Based on Social Force Swarm Optimization Algorithm in Cloud Computing
计算机科学, 2015, 42(4): 206-208. https://doi.org/10.11896/j.issn.1002-137X.2015.04.041
[13] 王群,戴秀岳.
CDN-P2P混合架构下直播流媒体系统服务应急保障研究
Research on Service Contingency Guarantee in CDN-P2P Hybrid Architecture Based Live Streaming
计算机科学, 2014, 41(Z6): 466-471.
[14] 储雅,马廷淮,赵立成.
云计算资源调度:策略与算法
Cloud Computing Resource Scheduling:Policy and Algorithm
计算机科学, 2013, 40(11): 8-13.
[15] 林伟伟,齐德呈.
云计算资源调度研究综述
Survey of Resource Scheduling in Cloud Computing
计算机科学, 2012, 39(10): 1-6.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!