High Performance Heterogeneous Multicore Architectures: A Study

Research output: Contribution to conference typesPaperResearchpeer-review

Abstract

The significant increase in the need for high- performance and energy-efficient computing systems has introduced heterogenous computing. However, the incorporation of different architectures into one system complicates the distribution of the workload between architectures. To address this challenge while meeting the goals of high-performance computing systems, several research contributions have been made. This paper reviews some of the proposed workload partitioning approaches for GPU-based, DSP-based and FPGA-based heterogenous systems. This research also covers some comparison studies regarding the FPGA versus DSP and FPGA versus GPU debates, showing that sometimes collaboration between these architectures seems to be the key. The aim of this study is to provide academic and industrial researchers with an insight of techniques to achieve the workload balancing in heterogenous systems and motivate them for further research in the field.
Original languageEnglish
Pages1-5
Number of pages5
Publication statusAccepted/In press - 25 May 2019
EventInternational Symposium on Computer Science and Intelligent Control - Netherlands, Amsterdam, Netherlands
Duration: 25 Sep 201927 Sep 2019
http://www.iscsic.org/

Conference

ConferenceInternational Symposium on Computer Science and Intelligent Control
Abbreviated titleISCSIC 2019
CountryNetherlands
CityAmsterdam
Period25/09/1927/09/19
Internet address

Fingerprint

Field programmable gate arrays (FPGA)
Graphics processing unit

Keywords

  • CPU
  • GPU
  • DSP
  • FPGA
  • Processing Unit
  • Workload Partitioning
  • Processing Capabilities

Cite this

Hoxha, I., & Opoku Agyeman, M. (Accepted/In press). High Performance Heterogeneous Multicore Architectures: A Study. 1-5. Paper presented at International Symposium on Computer Science and Intelligent Control, Amsterdam, Netherlands.
Hoxha, Igla ; Opoku Agyeman, Michael. / High Performance Heterogeneous Multicore Architectures: A Study. Paper presented at International Symposium on Computer Science and Intelligent Control, Amsterdam, Netherlands.5 p.
@conference{915154b80c984e359e5139f721a1bc4c,
title = "High Performance Heterogeneous Multicore Architectures: A Study",
abstract = "The significant increase in the need for high- performance and energy-efficient computing systems has introduced heterogenous computing. However, the incorporation of different architectures into one system complicates the distribution of the workload between architectures. To address this challenge while meeting the goals of high-performance computing systems, several research contributions have been made. This paper reviews some of the proposed workload partitioning approaches for GPU-based, DSP-based and FPGA-based heterogenous systems. This research also covers some comparison studies regarding the FPGA versus DSP and FPGA versus GPU debates, showing that sometimes collaboration between these architectures seems to be the key. The aim of this study is to provide academic and industrial researchers with an insight of techniques to achieve the workload balancing in heterogenous systems and motivate them for further research in the field.",
keywords = "CPU, GPU, DSP, FPGA, Processing Unit, Workload Partitioning, Processing Capabilities",
author = "Igla Hoxha and {Opoku Agyeman}, Michael",
year = "2019",
month = "5",
day = "25",
language = "English",
pages = "1--5",
note = "International Symposium on Computer Science and Intelligent Control, ISCSIC 2019 ; Conference date: 25-09-2019 Through 27-09-2019",
url = "http://www.iscsic.org/",

}

Hoxha, I & Opoku Agyeman, M 2019, 'High Performance Heterogeneous Multicore Architectures: A Study' Paper presented at International Symposium on Computer Science and Intelligent Control, Amsterdam, Netherlands, 25/09/19 - 27/09/19, pp. 1-5.

High Performance Heterogeneous Multicore Architectures: A Study. / Hoxha, Igla; Opoku Agyeman, Michael.

2019. 1-5 Paper presented at International Symposium on Computer Science and Intelligent Control, Amsterdam, Netherlands.

Research output: Contribution to conference typesPaperResearchpeer-review

TY - CONF

T1 - High Performance Heterogeneous Multicore Architectures: A Study

AU - Hoxha, Igla

AU - Opoku Agyeman, Michael

PY - 2019/5/25

Y1 - 2019/5/25

N2 - The significant increase in the need for high- performance and energy-efficient computing systems has introduced heterogenous computing. However, the incorporation of different architectures into one system complicates the distribution of the workload between architectures. To address this challenge while meeting the goals of high-performance computing systems, several research contributions have been made. This paper reviews some of the proposed workload partitioning approaches for GPU-based, DSP-based and FPGA-based heterogenous systems. This research also covers some comparison studies regarding the FPGA versus DSP and FPGA versus GPU debates, showing that sometimes collaboration between these architectures seems to be the key. The aim of this study is to provide academic and industrial researchers with an insight of techniques to achieve the workload balancing in heterogenous systems and motivate them for further research in the field.

AB - The significant increase in the need for high- performance and energy-efficient computing systems has introduced heterogenous computing. However, the incorporation of different architectures into one system complicates the distribution of the workload between architectures. To address this challenge while meeting the goals of high-performance computing systems, several research contributions have been made. This paper reviews some of the proposed workload partitioning approaches for GPU-based, DSP-based and FPGA-based heterogenous systems. This research also covers some comparison studies regarding the FPGA versus DSP and FPGA versus GPU debates, showing that sometimes collaboration between these architectures seems to be the key. The aim of this study is to provide academic and industrial researchers with an insight of techniques to achieve the workload balancing in heterogenous systems and motivate them for further research in the field.

KW - CPU

KW - GPU

KW - DSP

KW - FPGA

KW - Processing Unit

KW - Workload Partitioning

KW - Processing Capabilities

M3 - Paper

SP - 1

EP - 5

ER -

Hoxha I, Opoku Agyeman M. High Performance Heterogeneous Multicore Architectures: A Study. 2019. Paper presented at International Symposium on Computer Science and Intelligent Control, Amsterdam, Netherlands.