Mobile Computing, Applications, and Services. Third International Conference, MobiCASE 2011, Los Angeles, CA, USA, October 24-27, 2011. Revised Selected Papers

Research Article

A Parallel Approach to Mobile Web Browsing

Download
513 downloads
  • @INPROCEEDINGS{10.1007/978-3-642-32320-1_22,
        author={Kiho Kim and Hoon-Mo Yang and Cheong-Ghil Kim and Shin-Dug Kim},
        title={A Parallel Approach to Mobile Web Browsing},
        proceedings={Mobile Computing, Applications, and Services. Third International Conference, MobiCASE 2011, Los Angeles, CA, USA, October 24-27, 2011. Revised Selected Papers},
        proceedings_a={MOBICASE},
        year={2012},
        month={10},
        keywords={mobile web browser parallel algorithm multi-core tree traversal},
        doi={10.1007/978-3-642-32320-1_22}
    }
    
  • Kiho Kim
    Hoon-Mo Yang
    Cheong-Ghil Kim
    Shin-Dug Kim
    Year: 2012
    A Parallel Approach to Mobile Web Browsing
    MOBICASE
    Springer
    DOI: 10.1007/978-3-642-32320-1_22
Kiho Kim1,*, Hoon-Mo Yang1,*, Cheong-Ghil Kim1,*, Shin-Dug Kim1,*
  • 1: Yonsei university
*Contact email: heavyarms3@gmail.com, hmyang@yonsei.ac.kr, cgkim@nsu.ac.kr, sdkim@yonsei.ac.kr

Abstract

This paper present a parallel approach about mobile web browsing, especially layout and paint parts. Web browser is one of the most frequently used applications in mobile devices and performance of web browser is an important factor affecting mobile device user experience. From our previous research, we found that layout and paint takes significant portion of web browser execution time and has similar execution characteristics. In this paper, we propose parallel render tree traversal algorithm for layout and paint parts in web browser: creating thread for sub-tree traversal processing. Moreover, to validate proposed Algorithm, we design a simple simulation implementing parallel tree traversal with web page render tree. The experiment results show that execution time is reduced average 28% in dual-core, 32% in quad-core compare to single-thread execution in paint simulation. In layout simulation, average 38% in dual-core, 57% in quad-core execution time is reduced.