Web-enabled Cluster Computing with Java

碩士 === 國立清華大學 === 資訊科學研究所 === 85 ===   With the vast growth of World Wide Web (WWW), Internet has become more and more useful and popular. People can get all kinds of information from Internet through Web browsers. This trend has a great impact on our computing environments. Computations need not b...

Full description

Bibliographic Details
Main Authors: Chen, Hsi-Che, 陳希哲
Other Authors: King, Chung-Ta
Format: Others
Language:zh-TW
Published: 1997
Online Access:http://ndltd.ncl.edu.tw/handle/59979955721811829305
id ndltd-TW-085NTHU3394026
record_format oai_dc
spelling ndltd-TW-085NTHU33940262015-10-13T18:05:33Z http://ndltd.ncl.edu.tw/handle/59979955721811829305 Web-enabled Cluster Computing with Java 針對Java語言的Web-enabled網路叢集計算 Chen, Hsi-Che 陳希哲 碩士 國立清華大學 資訊科學研究所 85   With the vast growth of World Wide Web (WWW), Internet has become more and more useful and popular. People can get all kinds of information from Internet through Web browsers. This trend has a great impact on our computing environments. Computations need not be confined to local machines any longer. They may be stretched across multiple machines over the Internet. Therefore, how to model cooperative computing and take advantage of the computing power spread across the Internet has become a very important problem.   In this thesis, we propose a Web-enable cluster computing system. The system allows cluster-based computation resources to be shared by users all over the world through WWW. The Web-enable cluster computing system contains five components: a server Web interface, a Java PVM platform, an AOV-aware execution facility, a matrix package and a load balancer. The server Web interface with users to load their Java programs, sets up the desired execution environment, compiles the source code and distributes the executable image to other computers in the cluster. The Java PVM platform supports low-level communication between java processes and presents to the user Java programs a virtual distributed-memory multiprocessor on top of the cluster. Our cluster environment provides supports to facilitate the programming of systems representable in activity-on-vertex networks, It makes communication between processes in the AOV network easy. A matrix package is provided to improve the performance of matrix operations by taking advantage of cluster computing. Finally, the load balancer balances the load in the cluster by dispatching tasks to lightly-loaded computers.   A possible use of such a Web-enabled cluster computing environment is in organizations such as supercomputer centers in which users submit their jobs through the Internet and run their programs over the various computers, often concurrently, in the center. In this way, computing resources may be shared more efficiently. In addition, through Internet and Web-based browsers, the computing resources may be accessed more easily from remote even without user login first.   In this thesis, we will introduce the concept and discuss the issues involved in developing a Web-enabled cluster computing environment, and describe the implementation of such an environment. Its usage and performance will also be presented. King, Chung-Ta 金仲達 1997 學位論文 ; thesis 38 zh-TW
collection NDLTD
language zh-TW
format Others
sources NDLTD
description 碩士 === 國立清華大學 === 資訊科學研究所 === 85 ===   With the vast growth of World Wide Web (WWW), Internet has become more and more useful and popular. People can get all kinds of information from Internet through Web browsers. This trend has a great impact on our computing environments. Computations need not be confined to local machines any longer. They may be stretched across multiple machines over the Internet. Therefore, how to model cooperative computing and take advantage of the computing power spread across the Internet has become a very important problem.   In this thesis, we propose a Web-enable cluster computing system. The system allows cluster-based computation resources to be shared by users all over the world through WWW. The Web-enable cluster computing system contains five components: a server Web interface, a Java PVM platform, an AOV-aware execution facility, a matrix package and a load balancer. The server Web interface with users to load their Java programs, sets up the desired execution environment, compiles the source code and distributes the executable image to other computers in the cluster. The Java PVM platform supports low-level communication between java processes and presents to the user Java programs a virtual distributed-memory multiprocessor on top of the cluster. Our cluster environment provides supports to facilitate the programming of systems representable in activity-on-vertex networks, It makes communication between processes in the AOV network easy. A matrix package is provided to improve the performance of matrix operations by taking advantage of cluster computing. Finally, the load balancer balances the load in the cluster by dispatching tasks to lightly-loaded computers.   A possible use of such a Web-enabled cluster computing environment is in organizations such as supercomputer centers in which users submit their jobs through the Internet and run their programs over the various computers, often concurrently, in the center. In this way, computing resources may be shared more efficiently. In addition, through Internet and Web-based browsers, the computing resources may be accessed more easily from remote even without user login first.   In this thesis, we will introduce the concept and discuss the issues involved in developing a Web-enabled cluster computing environment, and describe the implementation of such an environment. Its usage and performance will also be presented.
author2 King, Chung-Ta
author_facet King, Chung-Ta
Chen, Hsi-Che
陳希哲
author Chen, Hsi-Che
陳希哲
spellingShingle Chen, Hsi-Che
陳希哲
Web-enabled Cluster Computing with Java
author_sort Chen, Hsi-Che
title Web-enabled Cluster Computing with Java
title_short Web-enabled Cluster Computing with Java
title_full Web-enabled Cluster Computing with Java
title_fullStr Web-enabled Cluster Computing with Java
title_full_unstemmed Web-enabled Cluster Computing with Java
title_sort web-enabled cluster computing with java
publishDate 1997
url http://ndltd.ncl.edu.tw/handle/59979955721811829305
work_keys_str_mv AT chenhsiche webenabledclustercomputingwithjava
AT chénxīzhé webenabledclustercomputingwithjava
AT chenhsiche zhēnduìjavayǔyándewebenabledwǎnglùcóngjíjìsuàn
AT chénxīzhé zhēnduìjavayǔyándewebenabledwǎnglùcóngjíjìsuàn
_version_ 1718028485356683264