research-article

Programmability and performance portability aspects of heterogeneous multi-/manycore systems

Authors:

Jesper Larsson Träff,

Sabri PllanaAuthors Info & Claims

DATE '12: Proceedings of the Conference on Design, Automation and Test in Europe

Pages 1403 - 1408

Published: 12 March 2012 Publication History

Get Access

Abstract

We discuss three complementary approaches that can provide both portability and an increased level of abstraction for the programming of heterogeneous multicore systems. Together, these approaches also support performance portability, as currently investigated in the EU FP7 project PEPPHER. In particular, we consider (1) a library-based approach, here represented by the integration of the SkePU C++ skeleton programming library with the StarPU runtime system for dynamic scheduling and dynamic selection of suitable execution units for parallel tasks; (2) a language-based approach, here represented by the Offload-C++ high-level language extensions and Offload compiler to generate platform-specific code; and (3) a component-based approach, specifically the PEPPHER component system for annotating user-level application components with performance metadata, thereby preparing them for performance-aware composition. We discuss the strengths and weaknesses of these approaches and show how they could complement each other in an integrational programming framework for heterogeneous multicore systems.

References

[1]

Cédric Augonnet, Samuel Thibault, Raymond Namyst, and Pierre-André Wacrenier. StarPU: A Unified Platform for Task Scheduling on Heterogeneous Multicore Architectures. Concurrency and Computation: Practice and Experience, Special Issue: Euro-Par 2009 23(2), Feb 2011.

Digital Library

Google Scholar

[2]

Siegfried Benkner, Sabri Pllana, Jesper L. Träff, Philippas Tsigas, Uwe Dolinsky, Cedric Augonnet, Beverly Bachmayer, Christoph W. Kessler, David Moloney, and Vitaly Osipov. PEPPHER: Efficient and Productive Usage of Hybrid Computing Systems. IEEE Micro 31(5), September/October 2011.

Digital Library

Google Scholar

[3]

Codeplay Software Ltd. Cache classes reduce code changes Offload C++ Knowledge Base Entry, http://offload.codeplay.com/kb/136.html, 2011

Google Scholar

[4]

Pete Cooper, Uwe Dolinsky, Alastair F. Donaldson, Andrew Richards, Colin Riley, and George Russell. Offload - Automating Code Migration to Heterogeneous Multicore Systems. 5th Int. Conf. on High Performance and Embedded Architectures and Compilers (HiPEAC'10), 2010, pp. 337--352

Digital Library

Google Scholar

[5]

Usman Dastgeer, Christoph Kessler, and Samuel Thibault. Flexible runtime support for efficient skeleton programming. Proc. ParCo-2011 conference, Ghent, Belgium, Sep. 2011, to appear.

Google Scholar

[6]

Usman Dastgeer, Lu Li, and Christoph Kessler. Performance-aware dynamic composition of applications for heterogeneous multicore systems with the PEPPHER Composition Tool. Proc. 16th Workshop on Compilers for Parallel Computers (CPC'12), Padova, Italy, January 2012.

Google Scholar

[7]

Johan Enmyren and Christoph Kessler: SkePU: A Multi-Backend Skeleton Programming Library for Multi-GPU Systems. Proc. 4th Int. Workshop on High-Level Parallel Programming and Applications (HLPP-2010), Baltimore, USA, Sep. 2010. ACM.

Digital Library

Google Scholar

[8]

B. Gary, Learning OpenCV: computer vision with the OpenCV library. O'Reilly USA, 2008.

Google Scholar

[9]

Christoph W. Kessler and Welf Löwe. Optimized composition of performance-aware parallel components. Concurrency and Computation: Practice and Experience, to appear. Published online in Wiley Online Library. Sep. 2011.

Crossref

Google Scholar

[10]

Erich Marth and Siegfried Benkner. Language support for pipelined applications on heterogeneous many-core architectures. In Workshop on Hybrid Multi-core Computing, in conjunction with HiPC, 2011.

Google Scholar

[11]

A. Munshi et al., The OpenCL Specification version 1.1, www.khronos.org/opencl, 2010.

Google Scholar

[12]

D. Quinlan, ROSE: Compiler Support for Object-Oriented Frameworks, Parallel Processing Letters, vol. 49, 2005.

Google Scholar

[13]

M. Sandrieser, S. Benkner, and S. Pllana. Improving programmability of heterogeneous many-core systems via explicit platform descriptions. In Proc. 4th Int. Workshop on Multicore Software Engineering (IWMSE'11), Hawaii, USA, May 2011, pp. 17--24, ACM.

Digital Library

Google Scholar

Cited By

View all

Viebke AMemeti SPllana SAbraham A(2019)CHAOSThe Journal of Supercomputing10.1007/s11227-017-1994-x75:1(197-227)Online publication date: 1-Jan-2019
https://dl.acm.org/doi/10.1007/s11227-017-1994-x
Memeti SPllana SBinotto AKołodziej JBrandic I(2019)Using meta-heuristics and machine learning for software optimization of parallel computing systemsComputing10.1007/s00607-018-0614-9101:8(893-936)Online publication date: 1-Aug-2019
https://dl.acm.org/doi/10.1007/s00607-018-0614-9
Soudris DPapadopoulos LKessler CKehagias DPapadopoulos ASeferlis PChatzigeorgiou AAmpatzoglou AThibault SNamyst RPleiter DGaydadjiev GBecker THaefele MMudge TPnevmatikatos D(2018)EXA2PRO programming environmentProceedings of the 18th International Conference on Embedded Computer Systems: Architectures, Modeling, and Simulation10.1145/3229631.3239369(202-209)Online publication date: 15-Jul-2018
https://dl.acm.org/doi/10.1145/3229631.3239369
Show More Cited By

Index Terms

Programmability and performance portability aspects of heterogeneous multi-/manycore systems
1. Computer systems organization
  1. Architectures
    1. Parallel architectures
2. Hardware

Recommendations

Manycore performance-portability: Kokkos multidimensional array library
A New Overview of the Trilinos Project --Part 1

Large, complex scientific and engineering application code have a significant investment in computational kernels to implement their mathematical models. Porting these computational kernels to the collection of modern manycore accelerator devices is a ...
On the Programmability and Performance of Heterogeneous Platforms
ICPADS '13: Proceedings of the 2013 International Conference on Parallel and Distributed Systems

General-purpose computing on an ever-broadening array of parallel devices has led to an increasingly complex and multi-dimensional landscape with respect to programmability and performance optimization. The growing diversity of parallel architectures ...
Towards achieving performance portability using directives for accelerators
WACCPD '16: Proceedings of the Third International Workshop on Accelerator Programming Using Directives

In this paper we explore the performance portability of directives provided by OpenMP 4 and OpenACC to program various types of node architectures with attached accelerators, both self-hosted multicore and offload multicore/GPU. Our goal is to examine ...

Comments

Information & Contributors

Information

Published In

DATE '12: Proceedings of the Conference on Design, Automation and Test in Europe

March 2012

1690 pages

ISBN:9783981080186

General Chairs:
Wolfgang Rosenstiel
University of Tuebingen and edacentrum, DE
,
Enrico Macii
Politecnico di Torino, IT

Publisher

EDA Consortium

San Jose, CA, United States

Publication History

Published: 12 March 2012

Check for updates

Qualifiers

Research-article

Conference

DATE '12

Sponsor:

EDAA
EDAC
SIGDA
The Russian Academy of Sciences

DATE '12: Design, Automation and Test in Europe

March 12 - 16, 2012

Dresden, Germany

Acceptance Rates

Overall Acceptance Rate 518 of 1,794 submissions, 29%

Contributors

Other Metrics

View Article Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

9
Total Citations
View Citations
87
Total Downloads

Downloads (Last 12 months)1
Downloads (Last 6 weeks)0

Reflects downloads up to 26 Jul 2024

Other Metrics

View Author Metrics

Citations

Cited By

View all

Viebke AMemeti SPllana SAbraham A(2019)CHAOSThe Journal of Supercomputing10.1007/s11227-017-1994-x75:1(197-227)Online publication date: 1-Jan-2019
https://dl.acm.org/doi/10.1007/s11227-017-1994-x
Memeti SPllana SBinotto AKołodziej JBrandic I(2019)Using meta-heuristics and machine learning for software optimization of parallel computing systemsComputing10.1007/s00607-018-0614-9101:8(893-936)Online publication date: 1-Aug-2019
https://dl.acm.org/doi/10.1007/s00607-018-0614-9
Soudris DPapadopoulos LKessler CKehagias DPapadopoulos ASeferlis PChatzigeorgiou AAmpatzoglou AThibault SNamyst RPleiter DGaydadjiev GBecker THaefele MMudge TPnevmatikatos D(2018)EXA2PRO programming environmentProceedings of the 18th International Conference on Embedded Computer Systems: Architectures, Modeling, and Simulation10.1145/3229631.3239369(202-209)Online publication date: 15-Jul-2018
https://dl.acm.org/doi/10.1145/3229631.3239369
Zhang WDuan PXie XXia FLu QLiu XZhou J(2016)QoS4IVSaaSPersonal and Ubiquitous Computing10.1007/s00779-016-0945-520:5(795-808)Online publication date: 1-Oct-2016
https://dl.acm.org/doi/10.1007/s00779-016-0945-5
Lee JSamadi MPark YMahlke S(2015)SKMDACM Transactions on Computer Systems10.1145/279872533:3(1-27)Online publication date: 31-Aug-2015
https://dl.acm.org/doi/10.1145/2798725
Benkner S(2015)Programming Support for Future Parallel ArchitecturesRevised Selected Papers of the 10th International Doctoral Workshop on Mathematical and Engineering Methods in Computer Science - Volume 954810.1007/978-3-319-29817-7_1(1-10)Online publication date: 23-Oct-2015
https://dl.acm.org/doi/10.1007/978-3-319-29817-7_1
McIntosh-Smith SCurran DMcIntosh-Smith SBergen B(2014)Evaluation of a performance portable lattice Boltzmann code using OpenCLProceedings of the International Workshop on OpenCL 2013 & 201410.1145/2664666.2664668(1-12)Online publication date: 12-May-2014
https://dl.acm.org/doi/10.1145/2664666.2664668
Fang SDu ZFang YHuang YChen YEeckhout LTemam OLi HChen YWu C(2014)Performance Portability Across Heterogeneous SoCs Using a Generalized Library-Based ApproachACM Transactions on Architecture and Code Optimization10.1145/260825311:2(1-25)Online publication date: 1-Jun-2014
https://dl.acm.org/doi/10.1145/2608253
Lee JSamadi MPark YMahlke SFensch CO'Boyle MSeznec ABodin F(2013)Transparent CPU-GPU collaboration for data-parallel kernels on heterogeneous systemsProceedings of the 22nd international conference on Parallel architectures and compilation techniques10.5555/2523721.2523756(245-256)Online publication date: 7-Oct-2013
https://dl.acm.org/doi/10.5555/2523721.2523756

View Options

Get Access

Login options

Check if you have access through your login credentials or your institution to get full access on this article.

Cited By

Index Terms

Recommendations

Manycore performance-portability: Kokkos multidimensional array library

On the Programmability and Performance of Heterogeneous Platforms

Towards achieving performance portability using directives for accelerators

Comments

Published In

Sponsors

Publisher

Publication History

Check for updates

Qualifiers

Conference

Acceptance Rates

Other Metrics

Article Metrics

Other Metrics

Cited By

Login options

Full Access

PDF

eReader

Abstract

References

Cited By

Index Terms

Recommendations

Manycore performance-portability: Kokkos multidimensional array library

On the Programmability and Performance of Heterogeneous Platforms

Towards achieving performance portability using directives for accelerators

Comments

Information

Published In

Sponsors

Publisher

Publication History

Check for updates

Qualifiers

Conference

Acceptance Rates

Contributors

Other Metrics

Bibliometrics

Article Metrics

Other Metrics

Citations

Cited By

Get Access

Login options

Full Access

View options

PDF

eReader

Figures

Other

Share

Share this Publication link

Share on social media

Affiliations