多租户集群中基于服务水平目标的调度机制

doi:10.11772/j.issn.1001-9081.2015.04.0944

计算机应用 ›› 2015, Vol. 35 ›› Issue (4): 944-949.DOI: 10.11772/j.issn.1001-9081.2015.04.0944

多租户集群中基于服务水平目标的调度机制

杜雄杰¹, 王旻², 汤学海^2,3, 张章²

1. 国家计算机网络应急技术处理协调中心, 北京 100029;
2. 中国科学院信息工程研究所, 北京 100093;
3. 中国科学院大学, 北京 100049

收稿日期:2014-11-13 修回日期:2015-01-13 出版日期:2015-04-10 发布日期:2015-04-08
通讯作者: 王旻
作者简介:杜雄杰(1986-),男,湖南双峰人,工程师,博士,主要研究方向:云计算、网络安全; 王旻(1989-),男,天津人,实习研究员,硕士,主要研究方向:分布式计算; 汤学海(1989-),男,湖南株洲人,硕士研究生,主要研究方向:云计算; 张章(1987-),男,湖北武汉人,研究实习员,硕士,主要研究方向:分布式系统。
基金资助:
国家自然科学基金资助项目(60903047);国家863计划项目(2012AA01A401);国家科技支撑计划项目(2012BAH37B02);中国科学院先导专项(XDA06030200)。

Scheduling mechanism based on service level objective in multi-tenant cluster

DU Xiongjie¹, WANG Min², TANG Xuehai^2,3, ZHANG Zhang²

1. National Computer Network Emergency Response Technical Team/Coordination Center of China, Beijing 100029, China;
2. Institute of Information Engineering, Chinese Academy of Sciences, Beijing 100093, China;
3. University of Chinese Academy of Sciences, Beijing 100049, China

Received:2014-11-13 Revised:2015-01-13 Online:2015-04-10 Published:2015-04-08

摘要/Abstract

摘要：

针对多租户集群中无法保证作业服务水平目标(SLO)的问题,提出了一种多租户场景下基于SLO的调度机制,其中包括优先调度算法和资源抢占算法。优先调度算法区别考虑超额使用资源的租户和未超额使用资源的租户,赋予后者的作业更高的优先级,在此前提下选择紧急度最高的作业,优先为其分配资源;资源抢占算法在资源受限的情况下,选择紧急度超过阈值的作业实施资源抢占,并根据租户的资源使用情况,在相应的运行作业范围内选择紧急度最低的作业,抢占其资源。实验结果表明,与现有保证公平的多租户调度器Capacity Scheduler相比,该调度机制可以在兼顾作业执行效率和租户间公平的前提下,显著提高作业的截止时间保证率,从而保证业务的服务水平目标。

关键词: 云计算, 多租户, 服务水平目标, 集群调度, 公平

Abstract:

A scheduling mechanism based on Service Level Objective (SLO) in multi-tenant cluster, including a preference scheduling algorithm and a resource preemption algorithm, was proposed to solve the problem of the inability to guarantee the SLOs of jobs in multi-tenant clusters. The preference scheduling algorithm considered the users who overused resources above their quota and the users who did not, then assigned a higher priority to the jobs of the latter users, under this condition, the job with highest priority was preferentially allocated resources. When the resources was limited, the resource preemption algorithm preempted the resource for the jobs whose urgency was above the threshold, and chose the jobs with the lowest urgency in the corresponding range of the running jobs according to the resource usages of tenants.The experimental results show that, compared with the current multi-tenant scheduler named Capacity Scheduler, the proposed mechanism can significantly improve the deadline guarantee rate of jobs and SLO with guaranteeing the job execution efficiency and the equity among tenants at the same time.

Key words: cloud computing, multi-tenant, Service Level Objective (SLO), cluster scheduling, fairness

中图分类号:

TP393

杜雄杰, 王旻, 汤学海, 张章. 多租户集群中基于服务水平目标的调度机制[J]. 计算机应用, 2015, 35(4): 944-949.

DU Xiongjie, WANG Min, TANG Xuehai, ZHANG Zhang. Scheduling mechanism based on service level objective in multi-tenant cluster[J]. Journal of Computer Applications, 2015, 35(4): 944-949.

参考文献

[1] DEAN J, SANJAY G. MapReduce: simplified data processing on large clusters [J]. Communications of the ACM,2008, 51(1): 107-113.
[2] ZAHARIA M, CHOWDHURY M, DAS T, et al. Resilient distributed datasets: a fault-tolerant abstraction for in-memory cluster computing [C]// Proceedings of the 9th USENIX Conference on Networked Systems Design and Implementation. Berkeley: USENIX, 2012:2.
[3] TOSHNIWAL A, TANEJA S, SHUKLA A, et al. Storm@ twitter[C]// SIGMOD 2014: Proceedings of the 2014 ACM SIGMOD International Conference on Management of data. New York: ACM Press, 2014: 147-156.
[4] LIN H, HAN Y. Performance management for multi-tenant Web applications [J]. Chinese Journal of Computers, 2010, 33(10): 1881-1895.(林海略, 韩燕波. 多租户应用的性能管理关键问题研究[J]. 计算机学报, 2010, 33(10): 1881-1895.)
[5] ZHENG Y, REN X, KONG L. Multi-tenant data migration strategy in SaaS platform [J]. Research Journal of Applied Sciences, Engineering and Technology,2014,7(12): 2421-2426.
[6] PAREKH S, GANDHI N, HELLERSTEIN J, et al. Using control theory to achieve service level objectives in performance management [J]. Real-time Systems, 2002, 23(1/2): 127-141.
[7] LIN C, LI Y, WAN J. Optimization approaches for QoS in computer network: a survey [J]. Chinese Journal of Computers, 2014, 31(6): 402-405.(林闯, 李寅, 万剑雄. 计算机网络服务质量优化方法研究综述[J]. 计算机学报,2011,34(1):1-14.)
[8] LIN W, QI D. Survey of research scheduling in cloud computing [J].Computer Science, 2012,39(10):1-6.(林伟伟,齐德昱.云计算资源调度研究综述[J].计算机科学,2012,39(10): 1-6.)
[9] SANDHOLM T, LAI K. Dynamic proportional share scheduling in Hadoop [C]// JSSPP 2010: Proceedings of the 15th International Conference on Job Scheduling Strategies for Parallel Processing, LNCS 6253. Berlin: Springer-Verlag, 2010:110-131.
[10] CAPIT N, COSTA G D, GEORGIOU Y, et al. A batch scheduler with high level components [C]// CCGRID 2005: Proceedings of the Fifth IEEE International Symposium on Cluster Computing and the Grid. Washington, DC: IEEE Computer Society, 2005, 2: 776-783.
[11] ZAHARIA M. Job scheduling with the fair and capacity schedulers [EB/OL]. [2013-10-10]. http://people.csail.mit.edu/matei/talks/2009/hadoop_summit_fair_scheduler.pdf.
[12] YANG G. Research of multi-tenant technology and resource management in cloud computing [D]. Beijing: Beijing University of Posts and Telecommunications, 2013.(杨光. 云计算的多租户技术与资源管理研究[D]. 北京:北京邮电大学, 2013.)
[13] BELLARE K, CURINO C, MACHANAVAJIHALA A, et al. WOO: a Scalable and multi-tenant platform for continuous knowledge base synthesis [J]. Proceedings of the VLDB Endowment, 2013,6(11): 1114-1125.
[14] FERGUSON A D, BODIK P, KANDULA S, et al. Jockey: guaranteed job latency in data parallel clusters [C]// Proceedings of the 7th ACM European Conference on Computer Systems. New York: ACM Press, 2012: 99-112.
[15] CHO B, RAHMAN M, CHAJED T, et al. Natjam: design and evaluation of eviction policies for supporting priorities and deadlines in mapreduce clusters [C]// Proceedings of the 4th Annual Symposium on Cloud Computing. New York: ACM Press, 2013: Article No. 6.
[16] LEVERICH J, KOZVYRAKIS C. Reconciling high server utilization and sub-millisecond quality-of-service [C]// Proceedings of the Ninth European Conference on Computer Systems. New York: ACM Press,2014: Article No.4.

多租户集群中基于服务水平目标的调度机制

Scheduling mechanism based on service level objective in multi-tenant cluster

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	陈家豪, 殷新春. 基于云雾计算的可追踪可撤销密文策略属性基加密方案[J]. 计算机应用, 2021, 41(6): 1611-1620.
[2]	葛丽娜, 胡雨谷, 张桂芬, 陈园园. 云计算环境基于客体属性匹配的逆向混合访问控制方案[J]. 计算机应用, 2021, 41(6): 1604-1610.
[3]	杨翎, 姜春茂. 基于三支决策的虚拟机节能迁移策略[J]. 计算机应用, 2021, 41(4): 990-998.
[4]	孙晓玲, 杨光, 沈焱萍, 杨秋格, 陈涛. 基于可拆分倒排索引的可搜索加密方案[J]. 《计算机应用》唯一官方网站, 2021, 41(11): 3288-3294.
[5]	吕佳玉, 竺智荣, 姚志强. 云计算环境下的双通道数据动态加密策略[J]. 计算机应用, 2020, 40(8): 2268-2273.
[6]	陈程军, 毛莺池, 王绎超. 基于激活-熵的分层迭代剪枝策略的CNN模型压缩[J]. 计算机应用, 2020, 40(5): 1260-1265.
[7]	郭曙杰, 李志华, 蔺凯青. 云环境下基于模糊隶属度的虚拟机放置算法[J]. 计算机应用, 2020, 40(5): 1374-1381.
[8]	许英鑫, 孙磊, 赵建成, 郭松辉. 基于蚁群优化算法的虚拟现场可编程门阵列部署策略[J]. 计算机应用, 2020, 40(3): 747-752.
[9]	王庆永, 毛莺池, 王绎超, 王龙宝. 基于多微云协作的计算任务卸载[J]. 计算机应用, 2020, 40(2): 328-334.
[10]	刘福鑫, 李劲巍, 王熠弘, 李琳. 基于Kubernetes的云原生海量数据存储系统设计与实现[J]. 计算机应用, 2020, 40(2): 547-552.
[11]	杨哂哂, 吴慧珍, 庄黎丽, 吕宏武. 基于Markov过程的IaaS系统可用性建模与分析方法[J]. 计算机应用, 2020, 40(10): 3013-3018.
[12]	李叶飞, 徐超, 许道强, 邹云峰, 张晓达, 钱柱中. 基于累计工作量的在线大数据分析作业调度算法[J]. 计算机应用, 2019, 39(8): 2431-2437.
[13]	林立, 熊金波, 肖如良, 林铭炜, 陈秀华. Gaming@Edge:基于边缘节点的低延迟云游戏系统[J]. 计算机应用, 2019, 39(7): 2001-2007.
[14]	徐雅斌, 彭宏恩. 基于需求预测的PaaS平台资源分配方法[J]. 计算机应用, 2019, 39(6): 1583-1588.
[15]	李启锐, 彭志平, 崔得龙, 何杰光. 容器云环境虚拟资源配置策略的优化[J]. 计算机应用, 2019, 39(3): 784-789.