第6期 王春凯，等：易变数据流的系统资源配置方法 ·1285· olus:an optimizer for distributed intra-node-parallel ing statistical techniques[JI.Proceedings of the VLDB en- streaming systems[C]//Proceedings of 2013 IEEE 29th In- dowment..2012.511):1555-1566 ternational Conference on Data Engineering.Brisbane, [21]AKDERE M.CETINTEMEL U.RIONDATO M.et al. Australia,2013:1280-1283. Learning-based query performance modeling and predic- [9]FU T Z J,DING Jianbing,MA R T B,et al.DRS:dynam- tion[C]//Proceedings of 2012 IEEE 28th International ic resource scheduling for real-time analytics over fast Conference on Data Engineering.Washington,DC,USA, streams[C]//Proceedings of 2015 IEEE 35th International 2012:390-401 Conference on Distributed Computing Systems.Columbus, [22]Kafka[EB/OL].[2019-04-20].http://kafka.apache.org/. OH.USA.2015:411-420 [23]SAX M J,CASTELLANOS M.Building a transparent [10]BITRAN G R.MORABITO R.State-of-the-art survey: batching layer for storm.HPL-2013-69[R].Palo Alto. open queueing networks:optimization and performance California,USA:HP Labs,2014. [24]JOHN G H,LANGLEY P.Estimating continuous distri- evaluation models for discrete manufa cturing systems[J]. Production and operations management,1996,5(2): butions in Bayesian classifiers[C]//Proceedings of the El- eventh Conference on Uncertainty in Artificial Intelli- 163-193. [11]ANIELLO L,BALDONI R,QUERZONI L.Adaptive on- gence.Montreal,Que,Canada,1995:338-345. [25]HOEFFDING W.Probability inequalities for sums of line scheduling in storm[C]//Proceedings of the 7th ACM bounded random variables[J].Journal of the American International Conference on Distributed Event-Based Sys- statistical association,1963,58(301):13-30 tems.Arlington,Texas,USA,2013:207-218. [26]OZA N C,RUSSELL S.Experimental comparisons of [12]KHOSHKBARFOROUSHHA A.RANJAN R.GAIRE R. online and batch versions of bagging and boosting[C]// et al.Resource usage estimation of data stream pro- Proceedings of the Seventh ACM SIGKDD International cessing workloads in datacenter clouds[J].arXiv: conference on Knowledge Discovery and Data Mining. 1501.07020.2015. San Francisco,California,USA,2001:359-364. [13]BISHOP C M.Mixture density networks[R].Birming- [27]AHA D W,KIBLER D,ALBERT M K.Instance-based ham,UK:Aston University,1994. learning algorithms[J].Machine learning,1991,6(1): [14]POGGI N,CARRERA D,CALL A,et al.ALOJA:a sys- 37-66. tematic study of Hadoop deployment variables to enable [28]HiBench[EB/OL].[2019-08-10].https://github.com/intel- automated characterization of cost-effectiveness[Cl//Pro- hadoop/HiBench/. ceedings of 2014 IEEE International Conference on Big [29]TPC-H.TPC-H is a decision support benchmark[EB/OL]. Data.Washington,DC,USA,2014:905-913. [2019-08-10].http://www.tpc.org/tpch. [15]Apache Hadoop[EB/OL].[2019-04-20].http://hadoop. 作者简介： apache.org/. 王春凯，男，1981年生，博士后 [16]BERRAL JL.POGGI N.CARRERA D.et al.ALOJA- 主要研究方向为数据流管理、知识融 ML:a framework for automating characterization and 合。曾主持和参与中国博士后科学基 knowledge discovery in hadoop deployments[Cl//Pro- 金项目、国家重点研发计划项目、国家 ceedings of the 21th ACM SIGKDD International Confer- 自然科学基金项目以及其他横向课题 ence on Knowledge Discovery and Data Mining.Sydney, 的研究。发表学术论文10余篇。 NSW,Australia,2015:1701-1710. [17]JAMSHIDI P,CASALE G.An uncertainty-aware ap- proach to optimal configuration of stream processing sys- 庄福振，男，1983年生.副研究 tems[C]//Proceedings of 2016 IEEE 24th International 员。主要研究方向为迁移学习、数据 Symposium on Modeling,Analysis and Simulation of 挖掘、机器学习。曾主持和参与国家 Computer and Telecommunication Systems.London,UK, 重点研发计划项目、国家”863”计划项 2016:39-48. 目、”973”子课题、国家自然科学基金 [18]VAN AKEN D,PAVLO A,GORDON G J,et al.Auto- 项目以及其他横向课题的研究。发表 学术论文40余篇。 matic database management system tuning through large- scale machine learning[C]//Proceedings of the 2017 ACM 史忠植，男，1941年生，研究员。 International Conference on Management of Data.Chica- 主要研究方向为智能科学、人工智能 go,llinois,.USA,2017:1009-1024. 机器学习、知识工程等。1979年、 [19]ABADI M,AGARWAL A,BARHAM P,et al.Tensor- 1998年、2001年均获中国科学院科技 Flow:large-scale machine learning on heterogeneous dis- 进步二等奖，1994年获中国科学院科 tributed systems[J].ar Xiv:1603.04467,2016. 技进步特等奖，2002年获国家科技进 [20]LI Jiexing.KONIG A C,NARASAYYA V,et al.Robust 步二等奖。发表学术论文400余篇， estimation of resource consumption for SQL queries us- 出版专著5部。olus: an optimizer for distributed intra-node-parallel streaming systems[C]//Proceedings of 2013 IEEE 29th In￾ternational Conference on Data Engineering. Brisbane, Australia, 2013: 1280-1283. FU T Z J, DING Jianbing, MA R T B, et al. DRS: dynam￾ic resource scheduling for real-time analytics over fast streams[C]//Proceedings of 2015 IEEE 35th International Conference on Distributed Computing Systems. Columbus, OH, USA, 2015: 411-420. [9] BITRAN G R, MORABITO R. State-of-the-art survey: open queueing networks: optimization and performance evaluation models for discrete manufa cturing systems[J]. Production and operations management, 1996, 5(2): 163–193. [10] ANIELLO L, BALDONI R, QUERZONI L. Adaptive on￾line scheduling in storm[C]//Proceedings of the 7th ACM International Conference on Distributed Event-Based Sys￾tems. Arlington, Texas, USA, 2013: 207-218. [11] KHOSHKBARFOROUSHHA A, RANJAN R, GAIRE R, et al. Resource usage estimation of data stream pro￾cessing workloads in datacenter clouds[J]. arXiv: 1501.07020, 2015. [12] BISHOP C M. Mixture density networks[R]. Birming￾ham, UK: Aston University, 1994. [13] POGGI N, CARRERA D, CALL A, et al. ALOJA: a sys￾tematic study of Hadoop deployment variables to enable automated characterization of cost-effectiveness[C]//Pro￾ceedings of 2014 IEEE International Conference on Big Data. Washington, DC, USA, 2014: 905-913. [14] Apache Hadoop[EB/OL].[2019-04-20]. http://hadoop. apache.org/. [15] BERRAL J L, POGGI N, CARRERA D, et al. ALOJA￾ML: a framework for automating characterization and knowledge discovery in hadoop deployments[C]//Pro￾ceedings of the 21th ACM SIGKDD International Confer￾ence on Knowledge Discovery and Data Mining. Sydney, NSW, Australia, 2015: 1701-1710. [16] JAMSHIDI P, CASALE G. An uncertainty-aware ap￾proach to optimal configuration of stream processing sys￾tems[C]//Proceedings of 2016 IEEE 24th International Symposium on Modeling, Analysis and Simulation of Computer and Telecommunication Systems. London, UK, 2016: 39-48. [17] VAN AKEN D, PAVLO A, GORDON G J, et al. Auto￾matic database management system tuning through large￾scale machine learning[C]//Proceedings of the 2017 ACM International Conference on Management of Data. Chica￾go, Illinois, USA, 2017: 1009-1024. [18] ABADI M, AGARWAL A, BARHAM P, et al. Tensor￾Flow: large-scale machine learning on heterogeneous dis￾tributed systems[J]. arXiv: 1603.04467, 2016. [19] LI Jiexing, KÖNIG A C, NARASAYYA V, et al. Robust estimation of resource consumption for SQL queries us- [20] ing statistical techniques[J]. Proceedings of the VLDB en￾dowment, 2012, 5(11): 1555–1566. AKDERE M, ÇETINTEMEL U, RIONDATO M, et al. Learning-based query performance modeling and predic￾tion[C]//Proceedings of 2012 IEEE 28th International Conference on Data Engineering. Washington, DC, USA, 2012: 390-401. [21] [22] Kafka[EB/OL].[2019-04-20]. http://kafka.apache.org/. SAX M J, CASTELLANOS M. Building a transparent batching layer for storm. HPL-2013-69[R]. Palo Alto, California, USA: HP Labs, 2014. [23] JOHN G H, LANGLEY P. Estimating continuous distri￾butions in Bayesian classifiers[C]//Proceedings of the El￾eventh Conference on Uncertainty in Artificial Intelli￾gence. Montréal, Qué, Canada, 1995: 338-345. [24] HOEFFDING W. Probability inequalities for sums of bounded random variables[J]. Journal of the American statistical association, 1963, 58(301): 13–30. [25] OZA N C, RUSSELL S. Experimental comparisons of online and batch versions of bagging and boosting[C]// Proceedings of the Seventh ACM SIGKDD International conference on Knowledge Discovery and Data Mining. San Francisco, California, USA, 2001: 359-364. [26] AHA D W, KIBLER D, ALBERT M K. Instance-based learning algorithms[J]. Machine learning, 1991, 6(1): 37–66. [27] HiBench[EB/OL].[2019-08-10]. https://github.com/intel￾hadoop/HiBench/. [28] TPC-H. TPC-H is a decision support benchmark[EB/OL]. [2019-08-10]. http://www.tpc.org/tpch. [29] 作者简介： 王春凯，男，1981 年生，博士后， 主要研究方向为数据流管理、知识融 合。曾主持和参与中国博士后科学基 金项目、国家重点研发计划项目、国家 自然科学基金项目以及其他横向课题 的研究。发表学术论文 10 余篇。 庄福振，男，1983 年生，副研究 员。主要研究方向为迁移学习、数据 挖掘、机器学习。曾主持和参与国家 重点研发计划项目、国家”863”计划项 目、”973”子课题、国家自然科学基金 项目以及其他横向课题的研究。发表 学术论文 40 余篇。 史忠植，男，1941 年生，研究员。 主要研究方向为智能科学、人工智能、 机器学习、知识工程等。197 9 年 、 1998 年、2001 年均获中国科学院科技 进步二等奖，1994 年获中国科学院科 技进步特等奖，2002 年获国家科技进 步二等奖。发表学术论文 400 余篇， 出版专著 5 部。 第 6 期 王春凯，等：易变数据流的系统资源配置方法 ·1285·