東華大學圖書館 |

Language: English

Help

回圖書館首頁

手機版館藏查詢

Back

Switch To: Labeled | MARC Mode | ISBD

Cloud Computing and Its Applications...

Kang, Cao.

Linked to FindBook

Google Book

Amazon

博客來

Cloud Computing and Its Applications in GIS.

Record Type:	Language materials, printed : Monograph/item
Title/Author:	Cloud Computing and Its Applications in GIS./
Author:	Kang, Cao.
Description:	93 p.
Notes:	Source: Dissertation Abstracts International, Volume: 72-07, Section: B, page: .
Contained By:	Dissertation Abstracts International72-07B.
Subject:	Geodesy. -
Online resource:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3455153
ISBN:	9781124623702

Cloud Computing and Its Applications in GIS.
Kang, Cao.

Cloud Computing and Its Applications in GIS. - 93 p.

Source: Dissertation Abstracts International, Volume: 72-07, Section: B, page: .

Thesis (Ph.D.)--Clark University, 2011.

Cloud computing is a novel computing paradigm that offers highly scalable and highly available distributed computing services. The objectives of this research are to: 1. analyze and understand cloud computing and its potential for GIS; 2. discover the feasibilities of migrating truly spatial GIS algorithms to distributed computing infrastructures; 3. explore a solution to host and serve large volumes of raster GIS data efficiently and speedily. These objectives thus form the basis for three professional articles.

ISBN: 9781124623702Subjects--Topical Terms:

550741
Geodesy.

Cloud Computing and Its Applications in GIS.
LDR:05461nam 2200325 4500 001 1402525
005 20111102140022.5
008 130515s2011 ||||||||||||||||| ||eng d
020 $a 9781124623702
035 $a (UMI)AAI3455153
035 $a AAI3455153
040 $a UMI $c UMI
100 1 $a Kang, Cao. $3 1681720
245 1 0 $a Cloud Computing and Its Applications in GIS.
300 $a 93 p.
500 $a Source: Dissertation Abstracts International, Volume: 72-07, Section: B, page: .
500 $a Adviser: J. Ronald Eastman.
502 $a Thesis (Ph.D.)--Clark University, 2011.
520 $a Cloud computing is a novel computing paradigm that offers highly scalable and highly available distributed computing services. The objectives of this research are to: 1. analyze and understand cloud computing and its potential for GIS; 2. discover the feasibilities of migrating truly spatial GIS algorithms to distributed computing infrastructures; 3. explore a solution to host and serve large volumes of raster GIS data efficiently and speedily. These objectives thus form the basis for three professional articles.
520 $a The first article is entitled "Cloud Computing and Its Applications in GIS". This paper introduces the concept, structure, and features of cloud computing. Features of cloud computing such as scalability, parallelization, and high availability make it a very capable computing paradigm. Unlike High Performance Computing (HPC), cloud computing uses inexpensive commodity computers. The uniform administration systems in cloud computing make it easier to use than GRID computing. Potential advantages of cloud-based GIS systems such as lower barrier to entry are consequently presented. Three cloud-based GIS system architectures are proposed: public cloud- based GIS systems, private cloud-based GIS systems and hybrid cloud-based GIS systems. Public cloud-based GIS systems provide the lowest entry barriers for users among these three architectures, but their advantages are offset by data security and privacy related issues. Private cloud-based GIS systems provide the best data protection, though they have the highest entry barriers. Hybrid cloud-based GIS systems provide a compromise between these extremes. The second article is entitled "A cloud computing algorithm for the calculation of Euclidian distance for raster GIS". Euclidean distance is a truly spatial GIS algorithm. Classical algorithms such as the pushbroom and growth ring techniques require computational propagation through the entire raster image, which makes it incompatible with the distributed nature of cloud computing. This paper presents a parallel Euclidean distance algorithm that works seamlessly with the distributed nature of cloud computing infrastructures. The mechanism of this algorithm is to subdivide a raster image into sub-images and wrap them with a one pixel deep edge layer of individually computed distance information. Each sub-image is then processed by a separate node, after which the resulting sub-images are reassembled into the final output. It is shown that while any rectangular sub-image shape can be used, those approximating squares are computationally optimal. This study also serves as a demonstration of this subdivide and layer-wrap strategy, which would enable the migration of many truly spatial GIS algorithms to cloud computing infrastructures. However, this research also indicates that certain spatial GIS algorithms such as cost distance cannot be migrated by adopting this mechanism, which presents significant challenges for the development of cloud-based GIS systems.
520 $a The third article is entitled "A Distributed Storage Schema for Cloud Computing based Raster GIS Systems". This paper proposes a NoSQL Database Management System (NDDBMS) based raster GIS data storage schema. NDDBMS has good scalability and is able to use distributed commodity computers, which make it superior to Relational Database Management Systems (RDBMS) in a cloud computing environment. In order to provide optimized data service performance, the proposed storage schema analyzes the nature of commonly used raster GIS data sets. It discriminates two categories of commonly used data sets, and then designs corresponding data storage models for both categories. As a result, the proposed storage schema is capable of hosting and serving enormous volumes of raster GIS data speedily and efficiently on cloud computing infrastructures. In addition, the scheme also takes advantage of the data compression characteristics of Quadtrees, thus promoting efficient data storage.
520 $a Through this assessment of cloud computing technology, the exploration of the challenges and solutions to the migration of GIS algorithms to cloud computing infrastructures, and the examination of strategies for serving large amounts of GIS data in a cloud computing infrastructure, this dissertation lends support to the feasibility of building a cloud-based GIS system. However, there are still challenges that need to be addressed before a full-scale functional cloud-based GIS system can be successfully implemented. (Abstract shortened by UMI.)
590 $a School code: 0048.
650 4 $a Geodesy. $3 550741
650 4 $a Remote Sensing. $3 1018559
650 4 $a Computer Science. $3 626642
690 $a 0370
690 $a 0799
690 $a 0984
710 2 $a Clark University. $3 1017690
773 0 $t Dissertation Abstracts International $g 72-07B.
790 1 0 $a Eastman, J. Ronald, $e advisor
790 $a 0048
791 $a Ph.D.
792 $a 2011
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3455153