Keio University, Graduate School of Media and Governance
MAUI Project
Ph.D. Dissertation

[ English | Japanese ]
Back to Index Page

ACADEMIC YEAR 2002
NAME SUGIURA, Kazunori
TITLE Practical Resource Adaptation for Portable Computer Environment
ABSTRACT

In this research, a method of disclosing enhanced ability in devices that are used in trend computation environment to the application and users in terms of efective monitoring and confguring mechanism is established. User oriented resource management mechanism named RMS (Resource Management Scheme) is inspired by the unique resource factor based on "energy consumption" and "effciency" of the trend device.

Devices used in current computing environment characterized as mobile, ubiquitous aware support with assisted network connection availability, has come up with technology evolution in two perspectives, (1)performance improvement, and (2)resource consumption economization. With much more penetration of computing support in our living culture, the design principle and implementation patterns adopted for legacy computing software environment will greatly change in a forthcoming environment, where a variety of devices with autonomous computational ability become ubiquitous and transparent. In such environment, where occasional environmental changes take in place rapidly, adaptive and collaborative steering of newly developed devices are required.

Development of RMS began with de.nition of four goals: (1) device awareness (2) application awareness (3) user awareness (4) developer awareness. Recon.guration of device status and e.ective adaptation mechanism with device in-dependency to users and applications are considered. Active device probing and controlling method is prepared to the application users, while legacy system design conceals its mechanism autonomously. Using RMS to forthcoming environment results e.cient usage of resources, resulting less power consumption with maximum e.ciency.

Implementation of RMS to practical computation environment focusing on client users, service demands, and complex of both are accomplished by analysing e.ectiveness based on the technological developing background on performance and resource consumption.

Deployment of contribution software environment based on RMS are done with some e.ective evaluation results. DVTS(Digital Video Transport System) and DVoD(Digital Video-on-Demand) are implemented in RMS architecture focusing on using processor and network resource adaptation mechanism to broadband streaming application massive network demonstration based on this system outcomes 79% improved battery lifetime with less power consumption, and e.ective adaptability in networks resulting jitter and packet-loss free communications. When adapting RMS to service environment, such as SOHO(Small O.ce Home O.ce) server, collaborative enhancement in devices to users economizes the resource consumption with required e.ciency.

The contribution of RMS will e.ectively reduce the resource consumption of computational environment, where our living are dependent to its existence thus limitation in generation of energy sources. Basically computational environment is based on a layered end to end model with each layer considering strict resource limitation. Best e.ort basis of device management in governance of an legacy operating system will be transformed by the collaboration mechanism established by the top-most layer(people and the society) to the bottom layer(device).

This dissertation proposes a new de.nition in terms of computational supported environment with collaboration of the resource inside and outside the computational environment.

CONTACT To obtain the whole paper, please contact;
uhyo@sfc.wide.ad.jp



MAUI Proejct
Last update:

Back to Project Home Page