Development and application of a hybrid programming environment on an ARM/DSP system for high performance computing

Gaurav Mitra; Jonathan Bohmann; Ian Lintault; Alistair P. Rendell

doi:10.1109/IPDPS.2018.00038

Development and application of a hybrid programming environment on an ARM/DSP system for high performance computing

Gaurav Mitra, Jonathan Bohmann, Ian Lintault, Alistair P. Rendell

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

3 Citations (Scopus)

Abstract

The nCore Brown-Dwarf system has a unique architecture where each node is comprised of two different low-power System-on-Chip (LPSoC) processors from Texas Instruments; the ARM/DSP Keystone II SoC and the DSP based Keystone I SoC. These LPSoC processors have, through use of the C66x multi-core DSP, been shown to be capable of running floating-point intensive HPC application codes. However, it is non-Trivial to run such codes across all processing elements of a node simultaneously. This paper demonstrates a hybrid programming environment that combines OpenMP, OpenCL and MPI to enable application execution across multiple Brown-Dwarf nodes. This environment is evaluated using two diverse application codes. The first is Level-3 BLAS matrix multiplication (GEMM), which is a standard HPC floating-point intensive benchmark. The second is a unique real-world scientific code for biostructure based drug design developed by the Southwest Research Institute called Rhodium. Performance and energy-efficiency of Rhodium is presented alongside comparisons with conventional x86 based HPC systems with attached accelerators. Results indicate that the Brown-Dwarf system remains competitive with contemporary systems for memory-bound computations.

Original language	English
Title of host publication	2018 IEEE International Parallel and Distributed Processing Symposium (IPDPS)
Place of Publication	Los Alamitos, California
Publisher	Institute of Electrical and Electronics Engineers
Pages	286-295
Number of pages	10
ISBN (Electronic)	9781538643686
ISBN (Print)	9781538643693
DOIs	https://doi.org/10.1109/IPDPS.2018.00038
Publication status	Published - 6 Aug 2018
Externally published	Yes
Event	32nd IEEE International Parallel and Distributed Processing Symposium, IPDPS 2018 - Vancouver, Canada Duration: 21 May 2018 → 25 May 2018

Publication series

Name	Proceedings - 2018 IEEE 32nd International Parallel and Distributed Processing Symposium, IPDPS 2018

Conference

Conference	32nd IEEE International Parallel and Distributed Processing Symposium, IPDPS 2018
Country/Territory	Canada
City	Vancouver
Period	21/05/18 → 25/05/18

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 7 Affordable and Clean Energy

Keywords

Accelerator
ARM
Brown Dwarf
C66x
Drug Design
DSP
Energy Efficiency
Keystone II
OpenCL
OpenMP
Protein Docking
Protein Engineering

Access to Document

10.1109/IPDPS.2018.00038Licence: In Copyright

Cite this

Mitra, G., Bohmann, J., Lintault, I., & Rendell, A. P. (2018). Development and application of a hybrid programming environment on an ARM/DSP system for high performance computing. In 2018 IEEE International Parallel and Distributed Processing Symposium (IPDPS) (pp. 286-295). Article 8425182 (Proceedings - 2018 IEEE 32nd International Parallel and Distributed Processing Symposium, IPDPS 2018). Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/IPDPS.2018.00038

Mitra, Gaurav ; Bohmann, Jonathan ; Lintault, Ian et al. / Development and application of a hybrid programming environment on an ARM/DSP system for high performance computing. 2018 IEEE International Parallel and Distributed Processing Symposium (IPDPS). Los Alamitos, California : Institute of Electrical and Electronics Engineers, 2018. pp. 286-295 (Proceedings - 2018 IEEE 32nd International Parallel and Distributed Processing Symposium, IPDPS 2018).

@inproceedings{8985615719634e298381c596544cb146,

title = "Development and application of a hybrid programming environment on an ARM/DSP system for high performance computing",

abstract = "The nCore Brown-Dwarf system has a unique architecture where each node is comprised of two different low-power System-on-Chip (LPSoC) processors from Texas Instruments; the ARM/DSP Keystone II SoC and the DSP based Keystone I SoC. These LPSoC processors have, through use of the C66x multi-core DSP, been shown to be capable of running floating-point intensive HPC application codes. However, it is non-Trivial to run such codes across all processing elements of a node simultaneously. This paper demonstrates a hybrid programming environment that combines OpenMP, OpenCL and MPI to enable application execution across multiple Brown-Dwarf nodes. This environment is evaluated using two diverse application codes. The first is Level-3 BLAS matrix multiplication (GEMM), which is a standard HPC floating-point intensive benchmark. The second is a unique real-world scientific code for biostructure based drug design developed by the Southwest Research Institute called Rhodium. Performance and energy-efficiency of Rhodium is presented alongside comparisons with conventional x86 based HPC systems with attached accelerators. Results indicate that the Brown-Dwarf system remains competitive with contemporary systems for memory-bound computations.",

keywords = "Accelerator, ARM, Brown Dwarf, C66x, Drug Design, DSP, Energy Efficiency, Keystone II, OpenCL, OpenMP, Protein Docking, Protein Engineering",

author = "Gaurav Mitra and Jonathan Bohmann and Ian Lintault and Rendell, {Alistair P.}",

year = "2018",

month = aug,

day = "6",

doi = "10.1109/IPDPS.2018.00038",

language = "English",

isbn = "9781538643693",

series = "Proceedings - 2018 IEEE 32nd International Parallel and Distributed Processing Symposium, IPDPS 2018",

publisher = "Institute of Electrical and Electronics Engineers",

pages = "286--295",

booktitle = "2018 IEEE International Parallel and Distributed Processing Symposium (IPDPS)",

address = "United States",

note = "32nd IEEE International Parallel and Distributed Processing Symposium, IPDPS 2018 ; Conference date: 21-05-2018 Through 25-05-2018",

}

Mitra, G, Bohmann, J, Lintault, I & Rendell, AP 2018, Development and application of a hybrid programming environment on an ARM/DSP system for high performance computing. in 2018 IEEE International Parallel and Distributed Processing Symposium (IPDPS)., 8425182, Proceedings - 2018 IEEE 32nd International Parallel and Distributed Processing Symposium, IPDPS 2018, Institute of Electrical and Electronics Engineers, Los Alamitos, California, pp. 286-295, 32nd IEEE International Parallel and Distributed Processing Symposium, IPDPS 2018, Vancouver, Canada, 21/05/18. https://doi.org/10.1109/IPDPS.2018.00038

Development and application of a hybrid programming environment on an ARM/DSP system for high performance computing. / Mitra, Gaurav; Bohmann, Jonathan; Lintault, Ian et al.
2018 IEEE International Parallel and Distributed Processing Symposium (IPDPS). Los Alamitos, California: Institute of Electrical and Electronics Engineers, 2018. p. 286-295 8425182 (Proceedings - 2018 IEEE 32nd International Parallel and Distributed Processing Symposium, IPDPS 2018).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Development and application of a hybrid programming environment on an ARM/DSP system for high performance computing

AU - Mitra, Gaurav

AU - Bohmann, Jonathan

AU - Lintault, Ian

AU - Rendell, Alistair P.

PY - 2018/8/6

Y1 - 2018/8/6

N2 - The nCore Brown-Dwarf system has a unique architecture where each node is comprised of two different low-power System-on-Chip (LPSoC) processors from Texas Instruments; the ARM/DSP Keystone II SoC and the DSP based Keystone I SoC. These LPSoC processors have, through use of the C66x multi-core DSP, been shown to be capable of running floating-point intensive HPC application codes. However, it is non-Trivial to run such codes across all processing elements of a node simultaneously. This paper demonstrates a hybrid programming environment that combines OpenMP, OpenCL and MPI to enable application execution across multiple Brown-Dwarf nodes. This environment is evaluated using two diverse application codes. The first is Level-3 BLAS matrix multiplication (GEMM), which is a standard HPC floating-point intensive benchmark. The second is a unique real-world scientific code for biostructure based drug design developed by the Southwest Research Institute called Rhodium. Performance and energy-efficiency of Rhodium is presented alongside comparisons with conventional x86 based HPC systems with attached accelerators. Results indicate that the Brown-Dwarf system remains competitive with contemporary systems for memory-bound computations.

AB - The nCore Brown-Dwarf system has a unique architecture where each node is comprised of two different low-power System-on-Chip (LPSoC) processors from Texas Instruments; the ARM/DSP Keystone II SoC and the DSP based Keystone I SoC. These LPSoC processors have, through use of the C66x multi-core DSP, been shown to be capable of running floating-point intensive HPC application codes. However, it is non-Trivial to run such codes across all processing elements of a node simultaneously. This paper demonstrates a hybrid programming environment that combines OpenMP, OpenCL and MPI to enable application execution across multiple Brown-Dwarf nodes. This environment is evaluated using two diverse application codes. The first is Level-3 BLAS matrix multiplication (GEMM), which is a standard HPC floating-point intensive benchmark. The second is a unique real-world scientific code for biostructure based drug design developed by the Southwest Research Institute called Rhodium. Performance and energy-efficiency of Rhodium is presented alongside comparisons with conventional x86 based HPC systems with attached accelerators. Results indicate that the Brown-Dwarf system remains competitive with contemporary systems for memory-bound computations.

KW - Accelerator

KW - ARM

KW - Brown Dwarf

KW - C66x

KW - Drug Design

KW - DSP

KW - Energy Efficiency

KW - Keystone II

KW - OpenCL

KW - OpenMP

KW - Protein Docking

KW - Protein Engineering

UR - http://www.scopus.com/inward/record.url?scp=85052209681&partnerID=8YFLogxK

U2 - 10.1109/IPDPS.2018.00038

DO - 10.1109/IPDPS.2018.00038

M3 - Conference contribution

AN - SCOPUS:85052209681

SN - 9781538643693

T3 - Proceedings - 2018 IEEE 32nd International Parallel and Distributed Processing Symposium, IPDPS 2018

SP - 286

EP - 295

BT - 2018 IEEE International Parallel and Distributed Processing Symposium (IPDPS)

PB - Institute of Electrical and Electronics Engineers

CY - Los Alamitos, California

T2 - 32nd IEEE International Parallel and Distributed Processing Symposium, IPDPS 2018

Y2 - 21 May 2018 through 25 May 2018

ER -

Mitra G, Bohmann J, Lintault I, Rendell AP. Development and application of a hybrid programming environment on an ARM/DSP system for high performance computing. In 2018 IEEE International Parallel and Distributed Processing Symposium (IPDPS). Los Alamitos, California: Institute of Electrical and Electronics Engineers. 2018. p. 286-295. 8425182. (Proceedings - 2018 IEEE 32nd International Parallel and Distributed Processing Symposium, IPDPS 2018). doi: 10.1109/IPDPS.2018.00038

Development and application of a hybrid programming environment on an ARM/DSP system for high performance computing

Abstract

Publication series

Conference

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Cite this