Wiley-WCMC-Helal-Turkistany-ThinClient-Jan-2007

Wiley-WCMC-Helal-Turkistany-ThinClient-Jan-2007 - 1...

Info iconThis preview shows pages 1–2. Sign up to view the full content.

View Full Document Right Arrow Icon
1 Abstract —The thin-client computing model has the potential to significantly increase the performance of mobile computing environments. By delivering any application through a single, small-footprint client (called a thin client ) implemented on a mobile device, it is possible to optimize application performance without the need for building wireless application gateways. We thus present two significant contributions in the area of wireless thin-client computing. Firstly, a mathematical performance model is derived for wireless thin-client system. This model identifies factors that affect the performance of the system, and supports derivation and analysis of adaptation strategies to maintain a user-specified quality of service (QoS). Secondly, a proxy-based adaptation framework is developed for wireless thin-client systems, which dynamically optimizes performance of a wireless thin client via dynamically discovered context. This is implemented with rule-based fuzzy logic that responds to variations in wireless link bandwidth and client processing power. Our fuzzy inference engine uses contextual data to dynamically optimize tradeoffs among different quality of service parameters offered to end users. Additionally, our adaptation framework uses highly scalable wavelet-based image coding to provide scalable QoS that can degrade gracefully. Our thin-client adaptation framework shields the user from ill effects of highly variable wireless network quality and mobile device resources. This improves performance of active applications, in which the display changes frequently. Further, active application behavior may produce high transmission latency for screen updates, which can adversely affect user perception of QoS, resulting in poor interactivity. We report measured adaptive performance under realistic mobile device and network conditions, for several different clients and servers. Index Terms — mobility adaptations, mobile computing models, thin-client model. I. INTRODUCTION HE concept of application-level adaptation, which was introduced by M. Satyanarayanan (surveyed in [Jing99]), allows mobile computing applications to select certain types of service over others, via tradeoff analysis based on application M. Al-Turkistany is with the Computer and Information Science and Engineering Department, University of Florida, Gainesville, FL 32611 USA (malturk@cise.ufl.edu). A. Helal is with the Computer and Information Science and Engineering Department, University of Florida, Gainesville, FL 32611 USA (helal@cise.ufl.edu). M. Schmalz is with the Computer and Information Science and Engineering Department, University of Florida, Gainesville, FL 32611 USA (mssz@cise.ufl.edu). semantics or knowledge. Salient quality-of-service (QoS)
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Image of page 2
This is the end of the preview. Sign up to access the rest of the document.

Page1 / 10

Wiley-WCMC-Helal-Turkistany-ThinClient-Jan-2007 - 1...

This preview shows document pages 1 - 2. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online