Tem_357 Harnessing the Power of Digital Transformation, Artificial Intelligence and Big Data Analytics with Parallel Computing

Hoe Min and Bong, Chin Wei and Jamal Ahmad Dargham (2020) Tem_357 Harnessing the Power of Digital Transformation, Artificial Intelligence and Big Data Analytics with Parallel Computing. TEST: Engineering and Management, 82. pp. 5657-5664. ISSN 0193 - 4120

Text Tem_357 Harnessing the Power of Digital Transformation, Artificial Intelligence and Big Data Analytics with Parallel Computing.pdf Restricted to Registered users only Download (153kB)

Abstract

Traditionally, 2D and especially 3D forward modeling and inversion of large geophysical datasets are performed on supercomputing clusters. This was due to the fact computing time taken by using PC was too time consuming. With the introduction of parallel computing, attempts have been made to perform computationally intensive tasks on PC or clusters of personal computers where the computing power was based on Central Processing Unit (CPU). It is further enhanced with Graphical Processing Unit (GPU) as the GPU has become affordable with the launch of GPU based computing devices. Therefore this paper presents a didactic concept in learning and applying parallel computing with the use of General Purpose Graphical Processing Unit (GPGPU) was carried out and perform preliminary testing in migrating existing sequential codes for solving initially 2D forward modeling of geophysical dataset. There are many challenges in performing these tasks mainly due to lack of some necessary development software tools, but the preliminary findings are promising. Traditionally, 2D and especially 3D forward modeling and inversion of large geophysical datasets are performed on supercomputing clusters. This was due to the fact computing time taken by using PC was too time consuming. With the introduction of parallel computing, attempts have been made to perform computationally intensive tasks on PC or clusters of personal computers where the computing power was based on Central Processing Unit (CPU). It is further enhanced with Graphical Processing Unit (GPU) as the GPU has become affordable with the launch of GPU based computing devices. Therefore this paper presents a didactic concept in learning and applying parallel computing with the use of General Purpose Graphical Processing Unit (GPGPU) was carried out and perform preliminary testing in migrating existing sequential codes for solving initially 2D forward modeling of geophysical dataset. There are many challenges in performing these tasks mainly due to lack of some necessary development software tools, but the preliminary findings are promising.Traditionally, 2D and especially 3D forward modeling and inversion of large geophysical datasets are performed on supercomputing clusters. This was due to the fact computing time taken by using PC was too time consuming. With the introduction of parallel computing, attempts have been made to perform computationally intensive tasks on PC or clusters of personal computers where the computing power was based on Central Processing Unit (CPU). It is further enhanced with Graphical Processing Unit (GPU) as the GPU has become affordable with the launch of GPU based computing devices. Therefore this paper presents a didactic concept in learning and applying parallel computing with the use of General Purpose Graphical Processing Unit (GPGPU) was carried out and perform preliminary testing in migrating existing sequential codes for solving initially 2D forward modeling of geophysical dataset. There are many challenges in performing these tasks mainly due to lack of some necessary development software tools, but the preliminary findings are promising.

Item Type:	Article
Keyword:	Geophysical Dataset, GPGPU, Jetson TX2, Parallel Computing
Subjects:	?? QA75 ?? T Technology > TK Electrical engineering. Electronics Nuclear engineering
Department:	FACULTY > Faculty of Engineering
Depositing User:	NORAINI LABUK -
Date Deposited:	08 Apr 2020 13:25
Last Modified:	28 Mar 2021 19:40
URI:	https://eprints.ums.edu.my/id/eprint/25273

Actions (login required)

View Item