Issue No. 01 - Jan.-Feb. (2017 vol. 19)
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/MCSE.2017.12
Jiaqing Dong , Tsinghua University
Hao Yin , Tsinghua University
Lyu Yongqiang , Tsinghua University
Hao Li , Tsinghua University
Wei Wang , Tsinghua University
Compared to desktop computers, smartphones frequently connect to the Internet via various applications through private protocols, resulting in severe difficulties in searching publicly shared information and in measuring network-layer performance. Most related work has focused on device-based software-assisted measurement, which suffers scalability as well as usability issues. In this article, the authors propose TAM, a transparent agent architecture to monitor applications in terms of both network performance and content. TAM sets up distributed virtual agents for target mobile applications and can enable parallel large-scale measurements through high-density computing resources, such as the cloud. TAM can also identify and measure the infrastructures that provide services to mobile applications. It does this without any built-in assistance from mobile devices, making TAM act as a transparent service in a scalable and usable manner. Experimental results of the prototype implementation verify the architecture.
Mobile applications, Mobile communication, Monitoring, Crawlers, Androids, Humanoid robots, Current measurement, Mobile computing, Heterogeneous networks, Pervasive computing, Network architecture
J. Dong, H. Yin, L. Yongqiang, H. Li and W. Wang, "TAM: A Transparent Agent Architecture for Measuring Mobile Applications," in Computing in Science & Engineering, vol. 19, no. 1, pp. 54-61, 2017.