Advertisement

A Dynamic Load Balancing Strategy Based on HAProxy and TCP Long Connection Multiplexing Technology

  • Wei LiEmail author
  • Jinwei Liang
  • Xiang Ma
  • Bo Qin
  • Bang Liu
Conference paper
Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 891)

Abstract

A load balancing strategy based on HAProxy and TCP long connection multiplexing technology is proposed to solve the waste of network resources caused by TCP requests in the data storage module of electrical energy management system. First, based on the load capacity and processing capacity of the cluster server, a server with the best performance is selected to establish a long connection. Then the proposed TCP long connection multiplexing technology is used to reuse different TCP requests to this long TCP connection. Experiments show that this strategy not only enhances the processing capacity of the server, but also dynamically selects the best server. Compared with the default load balancing strategy in HAProxy, it has higher feasibility and stability.

Keywords

Multiplexing of TCP Load balancing HAProxy 

Notes

Acknowledgements

This work was supported by Shaanxi Province Technical Innovation Guide Special Project (2018SJRG-G-03) and Shaanxi education department industrialization project (16JF024).

References

  1. 1.
    Zhou, L., Wang, F.: Study on data load balancing method of railway passenger ticket system. Railw. Transp. Econ. 40(5), 46–50 (2018). https://doi.org/10.16668/j.cnki.issn.1003-1421.2018.05.09Google Scholar
  2. 2.
    Liu, K.: HAProxy is used to realize the Web load balance of course selection system. Comput. Knowl. Technol. 7(1), 35–36 (2011). https://doi.org/10.3969/j.issn.1009-3044.2011.01.015Google Scholar
  3. 3.
    Wang, W., Hao, X., Duan, G., et al.: The multiplexing mechanism of concurrent TCP in satellite network. J. Cent. South Univ. (Sci. Technol.) 48(3), 712–720 (2017). https://doi.org/10.11817/j.issn.1672-7207.2017.03.020Google Scholar
  4. 4.
    Liu, M.: Factors influencing TCP transmission rate and optimization methods. China Comput. Commun. (8), 87–88 (2016).  https://doi.org/10.3969/j.issn.1003-9767.2016.08.041
  5. 5.
    Dou, L., Lu, X., Duan, H.: Design and implementation of multiplexed network model based on TCP. Appl. Res. Comput. (6), 245–247 (2006).  https://doi.org/10.3969/j.issn.1001-3695.2006.06.081
  6. 6.
    Zhou, S.: Research and implementation of TCP long connection multiplexing based on HAProxy. South China University of Technology (2011)Google Scholar
  7. 7.
    Liu, J., Xu, L., Zhang, W.: Load balancing algorithm based on dynamic feedback. Comput. Eng. Sci. 25(5), 65–68 (2003). https://doi.org/10.3969/j.issn.1007-130x.2003Google Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  • Wei Li
    • 1
    Email author
  • Jinwei Liang
    • 1
  • Xiang Ma
    • 1
  • Bo Qin
    • 1
  • Bang Liu
    • 1
  1. 1.Xi’an University of Posts and TelecommunicationsXi’anChina

Personalised recommendations