Automatic Mapping of Array Operations to Specific Architectures

Simon Andreas Frimann Lund; Mads Ruben Burgdorff Kristensen; Brian Vinter

doi:10.1109/WOLFHPC.2016.05

Automatic Mapping of Array Operations to Specific Architectures

Simon Andreas Frimann Lund, Mads Ruben Burgdorff Kristensen, Brian Vinter

eScience

Abstract

Array-oriented programming has been around for about thirty years and provides a fundamental abstraction for scientific computing. However, a wealth of popular programming languages in existence fail to provide convenient highlevel abstractions and exploit parallelism. One reason being that hardware is an ever-moving target.For this purpose, we introduce CAPE, a C-targeting Array Processing Engine, which manages the concerns of optimizing and parallelizing the execution of array operations. It is intended as a backend for new and existing languages and provides a portable runtime with a C-interface.The performance of the implementation is studied in relation to high-level implementations of a set of applications, kernels and synthetic benchmarks in Python/NumPy as well as lowlevel implementations in C/C++. We show the performance improvement over the high-productivity environment and how close the implementation is to handcrafted C/C++ code.

Original language	English
Title of host publication	2016 Sixth International Workshop on Domain-Specific Languages and High-Level Frameworks for High Performance Computing
Place of Publication	Salt Lake City, USA
Publisher	IEEE
Publication date	27 Jan 2017
Chapter	16689245
DOIs	https://doi.org/10.1109/WOLFHPC.2016.05
Publication status	Published - 27 Jan 2017

Access to Document

10.1109/WOLFHPC.2016.05

07836080(1)Final published version, 193 KB

Cite this

Automatic Mapping of Array Operations to Specific Architectures. / Lund, Simon Andreas Frimann; Kristensen, Mads Ruben Burgdorff; Vinter, Brian.

2016 Sixth International Workshop on Domain-Specific Languages and High-Level Frameworks for High Performance Computing. Salt Lake City, USA : IEEE, 2017.

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

@inproceedings{f225eba40fcb40ce82c76e2f55fc440d,

title = "Automatic Mapping of Array Operations to Specific Architectures",

abstract = "Array-oriented programming has been around for about thirty years and provides a fundamental abstraction for scientific computing. However, a wealth of popular programming languages in existence fail to provide convenient highlevel abstractions and exploit parallelism. One reason being that hardware is an ever-moving target.For this purpose, we introduce CAPE, a C-targeting Array Processing Engine, which manages the concerns of optimizing and parallelizing the execution of array operations. It is intended as a backend for new and existing languages and provides a portable runtime with a C-interface.The performance of the implementation is studied in relation to high-level implementations of a set of applications, kernels and synthetic benchmarks in Python/NumPy as well as lowlevel implementations in C/C++. We show the performance improvement over the high-productivity environment and how close the implementation is to handcrafted C/C++ code.",

author = "Lund, {Simon Andreas Frimann} and Kristensen, {Mads Ruben Burgdorff} and Brian Vinter",

year = "2017",

month = jan,

day = "27",

doi = "10.1109/WOLFHPC.2016.05",

language = "English",

booktitle = "2016 Sixth International Workshop on Domain-Specific Languages and High-Level Frameworks for High Performance Computing",

publisher = "IEEE",

}

TY - GEN

T1 - Automatic Mapping of Array Operations to Specific Architectures

AU - Lund, Simon Andreas Frimann

AU - Kristensen, Mads Ruben Burgdorff

AU - Vinter, Brian

PY - 2017/1/27

Y1 - 2017/1/27

N2 - Array-oriented programming has been around for about thirty years and provides a fundamental abstraction for scientific computing. However, a wealth of popular programming languages in existence fail to provide convenient highlevel abstractions and exploit parallelism. One reason being that hardware is an ever-moving target.For this purpose, we introduce CAPE, a C-targeting Array Processing Engine, which manages the concerns of optimizing and parallelizing the execution of array operations. It is intended as a backend for new and existing languages and provides a portable runtime with a C-interface.The performance of the implementation is studied in relation to high-level implementations of a set of applications, kernels and synthetic benchmarks in Python/NumPy as well as lowlevel implementations in C/C++. We show the performance improvement over the high-productivity environment and how close the implementation is to handcrafted C/C++ code.

AB - Array-oriented programming has been around for about thirty years and provides a fundamental abstraction for scientific computing. However, a wealth of popular programming languages in existence fail to provide convenient highlevel abstractions and exploit parallelism. One reason being that hardware is an ever-moving target.For this purpose, we introduce CAPE, a C-targeting Array Processing Engine, which manages the concerns of optimizing and parallelizing the execution of array operations. It is intended as a backend for new and existing languages and provides a portable runtime with a C-interface.The performance of the implementation is studied in relation to high-level implementations of a set of applications, kernels and synthetic benchmarks in Python/NumPy as well as lowlevel implementations in C/C++. We show the performance improvement over the high-productivity environment and how close the implementation is to handcrafted C/C++ code.

U2 - 10.1109/WOLFHPC.2016.05

DO - 10.1109/WOLFHPC.2016.05

M3 - Article in proceedings

BT - 2016 Sixth International Workshop on Domain-Specific Languages and High-Level Frameworks for High Performance Computing

PB - IEEE

CY - Salt Lake City, USA

ER -

Automatic Mapping of Array Operations to Specific Architectures

Abstract

Access to Document

Fingerprint

Cite this