Having your cake and eating it: Exploiting Python for programmer productivity and performance on micro-core architectures using ePython

Maurice Jamieson; Nick Brown; Sihang Liu

doi:10.25080/Majora-342d178e-00f

Having your cake and eating it: Exploiting Python for programmer productivity and performance on micro-core architectures using ePython

Maurice Jamieson, Nick Brown, Sihang Liu

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

Abstract / Description of output

Micro-core architectures combine many simple, low memory, low power computing cores together in a single package. These can be used as a co-processor or standalone but due to limited on-chip memory and esoteric nature of the hardware, writing efficient parallel codes for these chips is challenging. In this paper we discuss our very low memory implementation of Python, ePython, supporting the rapid development of parallel Python codes for these co-processors. An offload abstraction is introduced, where programmers decorate specific functions in their Python code, running under any Python interpreter on the host CPU, with the underlying technology then taking care of the low level data movement, scheduling and ePython execution on the micro-core co-processor. A benchmark solving Laplace's equation for diffusion via Jacobi iteration is used to explore the performance of ePython on three different micro-core architectures, and introduces work around native compilation for micro-cores and the performance advantages that this can provide.

Original language	English
Title of host publication	Proceedings of the 19th Python in Science Conference
Subtitle of host publication	SciPy 2020
Editors	Meghann Agarwal, Chris Calloway, Dillon Niederhut, David Shupe
Place of Publication	Austin, Texas
Pages	97 - 105
Number of pages	9
DOIs	https://doi.org/10.25080/Majora-342d178e-00f
Publication status	Published - 10 Jul 2020
Event	Scientific Computing with Python 2020 - Virtual conference Duration: 6 Jul 2020 → 12 Jul 2020 https://www.scipy2020.scipy.org/

Publication series

Name	Proceedings of the Python in Science Conferences
Volume	2020
ISSN (Electronic)	2575-9752

Conference

Conference	Scientific Computing with Python 2020
Abbreviated title	SciPy 2020
City	Virtual conference
Period	6/07/20 → 12/07/20
Internet address	https://www.scipy2020.scipy.org/

Keywords / Materials (for Non-textual outputs)

ePython
micro-cores
RISC-V
MicroBlaze
PicoRV32
Epiphany

Access to Document

10.25080/Majora-342d178e-00fLicence: Creative Commons: Attribution (CC-BY)

Having your cake_JAMIESON_DOA04072020_VOR_CC-BYFinal published version, 198 KBLicence: Creative Commons: Attribution (CC-BY)

http://conference.scipy.org/proceedings/scipy2020/pdfs/maurice_jamieson.pdfLicence: Creative Commons: Attribution (CC-BY)

Cite this

Jamieson, M., Brown, N., & Liu, S. (2020). Having your cake and eating it: Exploiting Python for programmer productivity and performance on micro-core architectures using ePython. In M. Agarwal, C. Calloway, D. Niederhut, & D. Shupe (Eds.), Proceedings of the 19th Python in Science Conference : SciPy 2020 (pp. 97 - 105). (Proceedings of the Python in Science Conferences; Vol. 2020).. https://doi.org/10.25080/Majora-342d178e-00f

Jamieson, Maurice ; Brown, Nick ; Liu, Sihang. / Having your cake and eating it: Exploiting Python for programmer productivity and performance on micro-core architectures using ePython. Proceedings of the 19th Python in Science Conference : SciPy 2020. editor / Meghann Agarwal ; Chris Calloway ; Dillon Niederhut ; David Shupe. Austin, Texas, 2020. pp. 97 - 105 (Proceedings of the Python in Science Conferences).

@inproceedings{86ff8f29faef4ce0a29b583c7f020172,

title = "Having your cake and eating it: Exploiting Python for programmer productivity and performance on micro-core architectures using ePython",

abstract = "Micro-core architectures combine many simple, low memory, low power computing cores together in a single package. These can be used as a co-processor or standalone but due to limited on-chip memory and esoteric nature of the hardware, writing efficient parallel codes for these chips is challenging. In this paper we discuss our very low memory implementation of Python, ePython, supporting the rapid development of parallel Python codes for these co-processors. An offload abstraction is introduced, where programmers decorate specific functions in their Python code, running under any Python interpreter on the host CPU, with the underlying technology then taking care of the low level data movement, scheduling and ePython execution on the micro-core co-processor. A benchmark solving Laplace's equation for diffusion via Jacobi iteration is used to explore the performance of ePython on three different micro-core architectures, and introduces work around native compilation for micro-cores and the performance advantages that this can provide.",

keywords = "ePython, micro-cores, RISC-V, MicroBlaze, PicoRV32, Epiphany",

author = "Maurice Jamieson and Nick Brown and Sihang Liu",

year = "2020",

month = jul,

day = "10",

doi = "10.25080/Majora-342d178e-00f",

language = "English",

series = "Proceedings of the Python in Science Conferences",

pages = "97 -- 105",

editor = "Meghann Agarwal and Chris Calloway and Dillon Niederhut and David Shupe",

booktitle = "Proceedings of the 19th Python in Science Conference",

note = "Scientific Computing with Python 2020, SciPy 2020 ; Conference date: 06-07-2020 Through 12-07-2020",

url = "https://www.scipy2020.scipy.org/",

}

Jamieson, M , Brown, N & Liu, S 2020, Having your cake and eating it: Exploiting Python for programmer productivity and performance on micro-core architectures using ePython. in M Agarwal, C Calloway, D Niederhut & D Shupe (eds), Proceedings of the 19th Python in Science Conference : SciPy 2020. Proceedings of the Python in Science Conferences, vol. 2020, Austin, Texas, pp. 97 - 105, Scientific Computing with Python 2020, Virtual conference, 6/07/20. https://doi.org/10.25080/Majora-342d178e-00f

Having your cake and eating it: Exploiting Python for programmer productivity and performance on micro-core architectures using ePython. / Jamieson, Maurice ; Brown, Nick; Liu, Sihang.
Proceedings of the 19th Python in Science Conference : SciPy 2020. ed. / Meghann Agarwal; Chris Calloway; Dillon Niederhut; David Shupe. Austin, Texas, 2020. p. 97 - 105 (Proceedings of the Python in Science Conferences; Vol. 2020).

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

TY - GEN

T1 - Having your cake and eating it: Exploiting Python for programmer productivity and performance on micro-core architectures using ePython

AU - Jamieson, Maurice

AU - Brown, Nick

AU - Liu, Sihang

PY - 2020/7/10

Y1 - 2020/7/10

N2 - Micro-core architectures combine many simple, low memory, low power computing cores together in a single package. These can be used as a co-processor or standalone but due to limited on-chip memory and esoteric nature of the hardware, writing efficient parallel codes for these chips is challenging. In this paper we discuss our very low memory implementation of Python, ePython, supporting the rapid development of parallel Python codes for these co-processors. An offload abstraction is introduced, where programmers decorate specific functions in their Python code, running under any Python interpreter on the host CPU, with the underlying technology then taking care of the low level data movement, scheduling and ePython execution on the micro-core co-processor. A benchmark solving Laplace's equation for diffusion via Jacobi iteration is used to explore the performance of ePython on three different micro-core architectures, and introduces work around native compilation for micro-cores and the performance advantages that this can provide.

AB - Micro-core architectures combine many simple, low memory, low power computing cores together in a single package. These can be used as a co-processor or standalone but due to limited on-chip memory and esoteric nature of the hardware, writing efficient parallel codes for these chips is challenging. In this paper we discuss our very low memory implementation of Python, ePython, supporting the rapid development of parallel Python codes for these co-processors. An offload abstraction is introduced, where programmers decorate specific functions in their Python code, running under any Python interpreter on the host CPU, with the underlying technology then taking care of the low level data movement, scheduling and ePython execution on the micro-core co-processor. A benchmark solving Laplace's equation for diffusion via Jacobi iteration is used to explore the performance of ePython on three different micro-core architectures, and introduces work around native compilation for micro-cores and the performance advantages that this can provide.

KW - ePython

KW - micro-cores

KW - RISC-V

KW - MicroBlaze

KW - PicoRV32

KW - Epiphany

U2 - 10.25080/Majora-342d178e-00f

DO - 10.25080/Majora-342d178e-00f

M3 - Conference contribution

T3 - Proceedings of the Python in Science Conferences

SP - 97

EP - 105

BT - Proceedings of the 19th Python in Science Conference

A2 - Agarwal, Meghann

A2 - Calloway, Chris

A2 - Niederhut, Dillon

A2 - Shupe, David

CY - Austin, Texas

T2 - Scientific Computing with Python 2020

Y2 - 6 July 2020 through 12 July 2020

ER -

Jamieson M , Brown N, Liu S. Having your cake and eating it: Exploiting Python for programmer productivity and performance on micro-core architectures using ePython. In Agarwal M, Calloway C, Niederhut D, Shupe D, editors, Proceedings of the 19th Python in Science Conference : SciPy 2020. Austin, Texas. 2020. p. 97 - 105. (Proceedings of the Python in Science Conferences). doi: 10.25080/Majora-342d178e-00f

Having your cake and eating it: Exploiting Python for programmer productivity and performance on micro-core architectures using ePython

Abstract / Description of output

Publication series

Conference

Keywords / Materials (for Non-textual outputs)

Access to Document

Fingerprint

Cite this