期刊文献+

科研大数据平台关键技术与实践 被引量:6

Key Technologies and Practice of Big Data Platform for Scientific Research
在线阅读 下载PDF
导出
摘要 首先,以高能物理领域数据处理为例,分析了支撑科学研究的大数据平台在存储和处理能力、传输和共享等方面的挑战,说明现有技术已经难以满足日益快速增长的数据处理需求。然后,给出了科研大数据平台的典型架构,并讨论科研大数据平台的关键技术,包括数据采集与清洗、数据存储、数据处理、数据传输、数据共享与安全等技术,同时介绍了各种关键技术的研究现状或者主流系统。最后,介绍了中国科学院高能物理研究所科研大数据开放平台的建设思路和实现框架,该平台试图解决目前大数据发展过程中面临的一些问题,如数据开放和跨领域融合不够、跨地域数据传输性能低等,从而激活数据价值,降低应用门槛。 This paper firstly discusses the challenges of big data platform for scientific research in the abilities of data storage, processing, transferring and sharing based on the example of data processing in the field of high en-ergy physics. Then the paper presents a typical structure of big data platform for scientific research and gives a brief review on some key technologies, including data acquisition, cleaning, storage, analysis, transfer, sharing and security. The popular systems and the current research situation of these key technologies are also described in the paper. Fi-nally, the concept and framework of the big data platform of High Energy Physics Institute of Chinese Academy of Sciences is introduced in this paper. This open platform is established as the solution of challenges/problems in big data evolution, for example, data acquisition, data transferring, and application interface.
作者 程耀东 陈刚
出处 《工程研究(跨学科视野中的工程)》 CSCD 2014年第3期266-274,共9页 JOURNAL OF ENGINEERING STUDIES
基金 国家自然科学基金项目(11205179)
关键词 大数据 数据存储 并行数据处理 开放平台 big data data storage parallel data process open platform
  • 相关文献

参考文献36

  • 1Worldwide LHC Computing Grid. Home [EB/OL]. (2013)[2013-08-30]. http://wlcg.web.cern.ch.
  • 2李国杰.大数据研究的科学价值[J].中国计算机学会通讯,2012,8(9):8-15.
  • 3Gantz J, Reinsel D. The Digital Universe in 2020: Big Data, Bigger Digital Shadows, and Biggest Growth in the Far East: United States [J/OL]. IDC iView: IDC Analyze the Future, 2012. http://www.emc.com/collateral/analystreports/idc-digital-universe-united-states.pdf.
  • 4CERN. The Large Hadron Collider [EB/OL]. (2013) [2013-08-30]. http://cern.ch/lhc.
  • 5IHEP. BESIII Experiment [EB/OL]. (2013)[2013-08-30]. http://bes3.ihep.ac.cn.
  • 6黄秋兰,朱随江,程耀东,陈刚.GRASS文件预留系统的设计与实现[J].核电子学与探测技术,2011,31(9):969-972. 被引量:3
  • 7汪璐,石京燕,程耀东.基于Lustre的BES集群存储系统[J].核电子学与探测技术,2010,30(12):1574-1578. 被引量:4
  • 8Deng Z Y, Li W D, Lin L, et al. Experience of BESIII Data Production with Local Cluster and Distributed Computing Model [J]. Journal of Physics: Conference Series, 2012, 396(3): 032031.
  • 9Mahout. Scalable machine learning and data mining [EB/OL]. [2013-08-30]. http://mahout.apache.org.
  • 10Wenaus T. Challenges of the LHC: Computing [R/OL]. 20^th Anniversary of the Winter Aspen Physics Conferences. (2005-02-19)[2013-08-30]. http://conferences. fnal.gov/ aspen05/talks/LH C-Computing - Wenaus. pdf.

二级参考文献4

  • 1陈志凌,程耀东,刘爱贵,陈刚.海量存储文件系统CastorFS的设计与实现[J].计算机工程与应用,2006,42(4):108-110. 被引量:2
  • 2Lustre File System:http://www.sun.com/software/products/lustre/.
  • 3IOzone Filesystem Benchmark:http://www.iozone.org/.
  • 4Choosing an I/O Scheduler for Red Hat(R) Enterprise Linux(R) 4 and the 2.6 Kernel.http://www.redhat.com/magazine/008jun05/features/schedulers/.

共引文献134

同被引文献86

引证文献6

二级引证文献34

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部