Mobile Computing Architecture Based on Linux Kernel

碩士 === 國立中山大學 === 電機工程學系研究所 === 102 === Recent cloud services from many vendors are booming in the data manipulation applica-tions. Due to the rapid development of wireless communication, cloud services can en-hance the high level of bound and support more aspect. In this paper, we present the d...

Full description

Bibliographic Details
Main Authors: Chih-Hung Kao, 高志宏
Other Authors: Jih-ching Chiu
Format: Others
Language:zh-TW
Published: 2014
Online Access:http://ndltd.ncl.edu.tw/handle/78515223386696637343
Description
Summary:碩士 === 國立中山大學 === 電機工程學系研究所 === 102 === Recent cloud services from many vendors are booming in the data manipulation applica-tions. Due to the rapid development of wireless communication, cloud services can en-hance the high level of bound and support more aspect. In this paper, we present the design of a cloud service environment that can flexi-bly and generically provide the distributed computing for mobile device. The architec-ture named as the GDCP (Generic Distributed Computing Platform). The GDCP has two major features. The one, it provides the platform for the users having the generic computer with more computing power to design their acceleration functions and reg-ister them at this platform for the application designers who need to design performance issued programs or mobile APPs. The others, the GDCP propose a distributed compu-ting mechanism for the application designer to construct the performance issued applica-tions. For embedding the GDCP environments to the generic computer, we go deep survey how to design the platform engines in the Linux kernel, and how to manage the computing resource according to the CPU usage. GDCP provide a novel distributed computing service for users. The design of GDCP can be joined by general computer, can reduce the cost of GDCP setting up. The environment for designing custom accelerated function and APPs can let the APPs de-velopers construct the performance issued APPs. By the GDCP, we can achieve the dis-tribution computing generically and flexibly.