Analysis of time performance on heteregenous cluster / Hanisah Shamsuddin

Sequential computing concept will always give time constraint problem to who are involve in complex computation like calculate and estimating weather forecast, rendering animation computation and many more in almost their daily works. Sequential computing also will lead to many others problem like h...

Full description

Saved in:

Bibliographic Details
Main Author:	Shamsuddin, Hanisah
Format:	Thesis
Language:	English
Published:	2009
Subjects:	Programming languages (Electronic computers) Neural networks (Computer science)
Online Access:	https://ir.uitm.edu.my/id/eprint/64653/1/64653.pdf
Tags:	Add Tag No Tags, Be the first to tag this record!

id	my-uitm-ir.64653
record_format	uketd_dc
spelling	my-uitm-ir.646532023-04-19T01:30:58Z Analysis of time performance on heteregenous cluster / Hanisah Shamsuddin 2009 Shamsuddin, Hanisah Programming languages (Electronic computers) Neural networks (Computer science) Sequential computing concept will always give time constraint problem to who are involve in complex computation like calculate and estimating weather forecast, rendering animation computation and many more in almost their daily works. Sequential computing also will lead to many others problem like high cost maintenance if system crash, performance problem and many more. Therefore, there are many parallel computing solution concept being introduced by developers and inventors. There are many real time simulation and advanced modeling concept being introduced for our benefit. However, there is less user can afford to have parallel supercomputers. So, low cost parallel computing solution that consists of few computer or PC being connected in one network can substitute supercomputers. In this research, there are few testing had been conducted to analyze the time performance in this heterogeneous cluster. 10 equations running being conducted in matrix multiplication calculation programming and 15 running for CPI calculation programming. In CPI calculation programming, it will compute the value of Pi by numerical integration in both parallel and sequential computing concept. Then, in matrix multiplication programming, it will compute the matrix calculation by 10 different 2x2 matrix size. It is expected that time of processing the calculation program in parallel will be much more less than sequential which means processing calculation or computation in parallel computing concept is faster that sequential computing concept 2009 Thesis https://ir.uitm.edu.my/id/eprint/64653/ https://ir.uitm.edu.my/id/eprint/64653/1/64653.pdf text en public masters Universiti Teknologi MARA (UiTM) Faculty of Computer and Mathematical Sciences -, Siti Arpah
institution	Universiti Teknologi MARA
collection	UiTM Institutional Repository
language	English
advisor	-, Siti Arpah
topic	Programming languages (Electronic computers) Neural networks (Computer science)
spellingShingle	Programming languages (Electronic computers) Neural networks (Computer science) Shamsuddin, Hanisah Analysis of time performance on heteregenous cluster / Hanisah Shamsuddin
description	Sequential computing concept will always give time constraint problem to who are involve in complex computation like calculate and estimating weather forecast, rendering animation computation and many more in almost their daily works. Sequential computing also will lead to many others problem like high cost maintenance if system crash, performance problem and many more. Therefore, there are many parallel computing solution concept being introduced by developers and inventors. There are many real time simulation and advanced modeling concept being introduced for our benefit. However, there is less user can afford to have parallel supercomputers. So, low cost parallel computing solution that consists of few computer or PC being connected in one network can substitute supercomputers. In this research, there are few testing had been conducted to analyze the time performance in this heterogeneous cluster. 10 equations running being conducted in matrix multiplication calculation programming and 15 running for CPI calculation programming. In CPI calculation programming, it will compute the value of Pi by numerical integration in both parallel and sequential computing concept. Then, in matrix multiplication programming, it will compute the matrix calculation by 10 different 2x2 matrix size. It is expected that time of processing the calculation program in parallel will be much more less than sequential which means processing calculation or computation in parallel computing concept is faster that sequential computing concept
format	Thesis
qualification_level	Master's degree
author	Shamsuddin, Hanisah
author_facet	Shamsuddin, Hanisah
author_sort	Shamsuddin, Hanisah
title	Analysis of time performance on heteregenous cluster / Hanisah Shamsuddin
title_short	Analysis of time performance on heteregenous cluster / Hanisah Shamsuddin
title_full	Analysis of time performance on heteregenous cluster / Hanisah Shamsuddin
title_fullStr	Analysis of time performance on heteregenous cluster / Hanisah Shamsuddin
title_full_unstemmed	Analysis of time performance on heteregenous cluster / Hanisah Shamsuddin
title_sort	analysis of time performance on heteregenous cluster / hanisah shamsuddin
granting_institution	Universiti Teknologi MARA (UiTM)
granting_department	Faculty of Computer and Mathematical Sciences
publishDate	2009
url	https://ir.uitm.edu.my/id/eprint/64653/1/64653.pdf
_version_	1783735505693704192

Analysis of time performance on heteregenous cluster / Hanisah Shamsuddin

Similar Items