期刊文献+

基于分布式管道模式的管道服务框架设计与实现

Design and Implementation of Pipeline Service Framework Based on Distributed Pipeline Mode
在线阅读 下载PDF
导出
摘要 集中化,合并任务的处理是分布式设计中的一个难点。由于不同任务中需要频繁的数据交换,且需对不同任务的结果进行合并,使其难以使用目前主流的分布式设计模型。分布式管道设计模式是一种高度解耦合的分布式设计模型,适合处理业务粒度较大,数据交换频发,IO瓶颈较高的任务。文章在分布式管道设计模式的基础上,参考服务发现机制实现了一个高性能的管道服务框架。 Centralization and combined task processing is a difficulty in distributed design.Due to there are frequent data exchanges in different tasks and the results of different tasks need to be merged,it is difficult to use the current mainstream distributed design model.Distributed pipeline design pattern is a high decoupling distributed design model,which is suitable for the tasks with large business granularity,frequent data exchange and high IO bottleneck.Based on the distributed pipeline design pattern,this paper implements a high performance pipeline service framework with the reference of service discovery mechanism.
作者 戴振邦 江恩杰 刘力嘉 甘江伟 DAI Zhenbang;JIANG Enjie;LIU Lijia;GAN Jiangwei(School of Chemical Engineering,Northwest Minzu University,Lanzhou 730124,China;Information Engineering Department,Chengyi University College,Jimei University,Xiamen 361021,China;College of Physics and Electronic Information Engineering,Minjiang University,Fuzhou 350108,China)
出处 《现代信息科技》 2021年第7期44-49,53,共7页 Modern Information Technology
基金 大学生创新创业训练计划项目(省级)(202013471013)。
关键词 分布式 服务发现 动态负载均衡 管道任务 高解耦 distributed service discovery dynamic load balancing pipeline task high decoupling
作者简介 戴振邦(2000-),男,汉族,福建莆田人,本科在读,研究方向:分布式、数据挖掘、集群计算;江恩杰(2000-),男,汉族,福建漳州人,本科在读,研究方向:人工智能、模式识别、云服务;刘力嘉:(2000-),男,汉族,福建漳州人,本科在读,研究方向:虚拟化;甘江伟(2000-),男,汉族,福建永安人,本科在读,研究方向:微电子。
  • 相关文献

参考文献7

二级参考文献58

  • 1GORMAN M. CodeViz:a call graph generation utility for C/C + + [ EB/OL]. http://www, skynet, ie/- mel/projects/codeviz/.
  • 2GUSTAFSSON A. Egypt-create call graph from GCC RTL dump[ EB/ OL]. (2013). http ://www. gson. org/egypt/egypt, html.
  • 3LATTNER C. Introduction to the LLVM compiler infrastructure [ C]//Proc of Itanium Conference and Expo. 2006.
  • 4CHAN S C , GAO G R, CHAPMAN B. Open64 compiler infrastruc- ture for emerging mulficore/manycore architecture[ C ]//Proe of IEEE International Symposium on Parallel and Distributed Processing. [ S. 1. ] : IEEE Press, 2008 : 1.
  • 5JACOB B, LARSON P, LEITAO B, et al. SystemTap: instrumenting the Linux kernel for analyzing performance and functional problems [K]. [S. 1. ]:IBM Redbook, 2008.
  • 6GRAHAM S L, KESSLER P B, MCHUSICK M K. Gprof: a call graph execution profiler [ J]. ACM SIGPLAN Notices, 1982, 17 (6) :120-126.
  • 7WEIDENDORFER J. Sequential performance analysis with callgrind and Kcachegrind [ M ]//Tools for High Performance Computing. Ber- lin: Springer, 2008 : 93-113,.
  • 8ROSTEDT S. Finding origins of latencies using Ftraee [ EB/OL ]. ( 2009 ). http ://lwn. net/images/conf/rtlwsl 1/papers/proe/p02. pdf.
  • 9KORANNE S. Handbook of open source tools[ M]. [ S. 1. ] :Spring- er, 2010.
  • 10KENISTON J, PANCHAMUKHI P, HIRAMATSU M. Kernel probes (Kprobes) [ EB/OL]. http://wenku, it168, com/d 000298233. shtml.

共引文献90

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