MIMD Programs Execution Support on SIMD Machines: A Holistic Survey
The Single Instruction Multiple Data (SIMD) architecture, supported by various high-performance computing platforms, efficiently utilizes data-level parallelism. The SIMD model is used in traditional CPUs, dedicated vector systems, and accelerators such as GPUs, vector extensions, and Xeon Phi. It p...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
IEEE
2024-01-01
|
Series: | IEEE Access |
Subjects: | |
Online Access: | https://ieeexplore.ieee.org/document/10458910/ |
_version_ | 1797243400571846656 |
---|---|
author | Dheya Mustafa Ruba Alkhasawneh Fadi Obeidat Ahmed S. Shatnawi |
author_facet | Dheya Mustafa Ruba Alkhasawneh Fadi Obeidat Ahmed S. Shatnawi |
author_sort | Dheya Mustafa |
collection | DOAJ |
description | The Single Instruction Multiple Data (SIMD) architecture, supported by various high-performance computing platforms, efficiently utilizes data-level parallelism. The SIMD model is used in traditional CPUs, dedicated vector systems, and accelerators such as GPUs, vector extensions, and Xeon Phi. It provides performance throughput in computation-intensive and data-parallel applications. Despite the similarity of data-processing principles between these architectures, porting various programming models between the reviewed platforms is challenging. Furthermore, enhancing the programmability of these architectures is an important feature for utilizing their emerging computing power and simplifying programming complexity. This paper reviews the basic principles of optimization techniques to run asynchronous Multiple Instruction Multiple Data (MIMD) on SIMD accelerators. It also surveys several GPU programming paradigms and application programming interfaces (APIs) and classifies these frameworks into different groups based on their criteria. In addition, a review of studies that performed a comparison of the collaborative execution of GPUs with CPUs and Xeon Phi is presented in this paper. This study will be beneficial for developers and researchers in the field of computer architecture and parallel computing of intensive scientific applications, specifically for early-stage high-performance computing researchers, to obtain a brief overview of performance optimization opportunities as well as the challenges of existing SIMD platforms. |
first_indexed | 2024-04-24T18:54:31Z |
format | Article |
id | doaj.art-e2f812d1195041239c538012d680ae65 |
institution | Directory Open Access Journal |
issn | 2169-3536 |
language | English |
last_indexed | 2024-04-24T18:54:31Z |
publishDate | 2024-01-01 |
publisher | IEEE |
record_format | Article |
series | IEEE Access |
spelling | doaj.art-e2f812d1195041239c538012d680ae652024-03-26T17:46:47ZengIEEEIEEE Access2169-35362024-01-0112343543437710.1109/ACCESS.2024.337299010458910MIMD Programs Execution Support on SIMD Machines: A Holistic SurveyDheya Mustafa0https://orcid.org/0000-0003-1456-7377Ruba Alkhasawneh1https://orcid.org/0000-0001-9535-2612Fadi Obeidat2https://orcid.org/0000-0002-8731-0989Ahmed S. Shatnawi3https://orcid.org/0000-0002-6239-3298Department of Computer Engineering, Faculty of Engineering, The Hashemite University, Zarqa, JordanDepartment of Communication and Computer Engineering, Faculty of Engineering, Al-Ahliyya Amman University, Amman, JordanSynopsys Inc., Austin, TX, USADepartment of Software Engineering, Jordan University of Science and Technology, Irbid, JordanThe Single Instruction Multiple Data (SIMD) architecture, supported by various high-performance computing platforms, efficiently utilizes data-level parallelism. The SIMD model is used in traditional CPUs, dedicated vector systems, and accelerators such as GPUs, vector extensions, and Xeon Phi. It provides performance throughput in computation-intensive and data-parallel applications. Despite the similarity of data-processing principles between these architectures, porting various programming models between the reviewed platforms is challenging. Furthermore, enhancing the programmability of these architectures is an important feature for utilizing their emerging computing power and simplifying programming complexity. This paper reviews the basic principles of optimization techniques to run asynchronous Multiple Instruction Multiple Data (MIMD) on SIMD accelerators. It also surveys several GPU programming paradigms and application programming interfaces (APIs) and classifies these frameworks into different groups based on their criteria. In addition, a review of studies that performed a comparison of the collaborative execution of GPUs with CPUs and Xeon Phi is presented in this paper. This study will be beneficial for developers and researchers in the field of computer architecture and parallel computing of intensive scientific applications, specifically for early-stage high-performance computing researchers, to obtain a brief overview of performance optimization opportunities as well as the challenges of existing SIMD platforms.https://ieeexplore.ieee.org/document/10458910/Acceleratorsasynchronous applicationsGPUsirregular applicationsSIMD |
spellingShingle | Dheya Mustafa Ruba Alkhasawneh Fadi Obeidat Ahmed S. Shatnawi MIMD Programs Execution Support on SIMD Machines: A Holistic Survey IEEE Access Accelerators asynchronous applications GPUs irregular applications SIMD |
title | MIMD Programs Execution Support on SIMD Machines: A Holistic Survey |
title_full | MIMD Programs Execution Support on SIMD Machines: A Holistic Survey |
title_fullStr | MIMD Programs Execution Support on SIMD Machines: A Holistic Survey |
title_full_unstemmed | MIMD Programs Execution Support on SIMD Machines: A Holistic Survey |
title_short | MIMD Programs Execution Support on SIMD Machines: A Holistic Survey |
title_sort | mimd programs execution support on simd machines a holistic survey |
topic | Accelerators asynchronous applications GPUs irregular applications SIMD |
url | https://ieeexplore.ieee.org/document/10458910/ |
work_keys_str_mv | AT dheyamustafa mimdprogramsexecutionsupportonsimdmachinesaholisticsurvey AT rubaalkhasawneh mimdprogramsexecutionsupportonsimdmachinesaholisticsurvey AT fadiobeidat mimdprogramsexecutionsupportonsimdmachinesaholisticsurvey AT ahmedsshatnawi mimdprogramsexecutionsupportonsimdmachinesaholisticsurvey |