Accelerator support in the charm++ parallel programming model

Laxmikant V Kale, David M. Kunzman, Lukasz Wesolowski

Research output: Chapter in Book/Report/Conference proceedingChapter

Abstract

The increased popularity of accelerators and heterogeneous systems has naturally lead to their adoption in the realm of high performance computing (HPC). As the name implies, HPC applications are always striving to achieve greater performance, typically to perform scientific simulations, financial predictions, etc. The high peak computation rates (GFlops) of accelerators make them particularly attractive to programmers creating these compute intensive HPC codes. Charm++ is one of several parallel programming models used to develop HPC applications. As we will discuss, we believe that some features of the Charm++ programming model make it well suited for programming heterogeneous clusters as well as clusters with accelerators. We also describe the various extensions that have been introduced into the Charm++ programming model to add support for accelerators and heterogeneous clusters.

Original languageEnglish (US)
Title of host publicationScientific Computing with Multicore and Accelerators
PublisherCRC Press
Pages393-412
Number of pages20
ISBN (Electronic)9781439825372
ISBN (Print)9781439825365
DOIs
StatePublished - Jan 1 2010

Fingerprint

Parallel programming
Parallel Programming
Accelerator
Programming Model
Particle accelerators
High Performance
Computing
Heterogeneous Systems
Programming
Imply
Prediction
Simulation

ASJC Scopus subject areas

  • Computer Science(all)
  • Mathematics(all)

Cite this

Kale, L. V., Kunzman, D. M., & Wesolowski, L. (2010). Accelerator support in the charm++ parallel programming model. In Scientific Computing with Multicore and Accelerators (pp. 393-412). CRC Press. https://doi.org/10.1201/b10376

Accelerator support in the charm++ parallel programming model. / Kale, Laxmikant V; Kunzman, David M.; Wesolowski, Lukasz.

Scientific Computing with Multicore and Accelerators. CRC Press, 2010. p. 393-412.

Research output: Chapter in Book/Report/Conference proceedingChapter

Kale, LV, Kunzman, DM & Wesolowski, L 2010, Accelerator support in the charm++ parallel programming model. in Scientific Computing with Multicore and Accelerators. CRC Press, pp. 393-412. https://doi.org/10.1201/b10376
Kale LV, Kunzman DM, Wesolowski L. Accelerator support in the charm++ parallel programming model. In Scientific Computing with Multicore and Accelerators. CRC Press. 2010. p. 393-412 https://doi.org/10.1201/b10376
Kale, Laxmikant V ; Kunzman, David M. ; Wesolowski, Lukasz. / Accelerator support in the charm++ parallel programming model. Scientific Computing with Multicore and Accelerators. CRC Press, 2010. pp. 393-412
@inbook{2c2a2225c03d430baac6ce5659a16532,
title = "Accelerator support in the charm++ parallel programming model",
abstract = "The increased popularity of accelerators and heterogeneous systems has naturally lead to their adoption in the realm of high performance computing (HPC). As the name implies, HPC applications are always striving to achieve greater performance, typically to perform scientific simulations, financial predictions, etc. The high peak computation rates (GFlops) of accelerators make them particularly attractive to programmers creating these compute intensive HPC codes. Charm++ is one of several parallel programming models used to develop HPC applications. As we will discuss, we believe that some features of the Charm++ programming model make it well suited for programming heterogeneous clusters as well as clusters with accelerators. We also describe the various extensions that have been introduced into the Charm++ programming model to add support for accelerators and heterogeneous clusters.",
author = "Kale, {Laxmikant V} and Kunzman, {David M.} and Lukasz Wesolowski",
year = "2010",
month = "1",
day = "1",
doi = "10.1201/b10376",
language = "English (US)",
isbn = "9781439825365",
pages = "393--412",
booktitle = "Scientific Computing with Multicore and Accelerators",
publisher = "CRC Press",

}

TY - CHAP

T1 - Accelerator support in the charm++ parallel programming model

AU - Kale, Laxmikant V

AU - Kunzman, David M.

AU - Wesolowski, Lukasz

PY - 2010/1/1

Y1 - 2010/1/1

N2 - The increased popularity of accelerators and heterogeneous systems has naturally lead to their adoption in the realm of high performance computing (HPC). As the name implies, HPC applications are always striving to achieve greater performance, typically to perform scientific simulations, financial predictions, etc. The high peak computation rates (GFlops) of accelerators make them particularly attractive to programmers creating these compute intensive HPC codes. Charm++ is one of several parallel programming models used to develop HPC applications. As we will discuss, we believe that some features of the Charm++ programming model make it well suited for programming heterogeneous clusters as well as clusters with accelerators. We also describe the various extensions that have been introduced into the Charm++ programming model to add support for accelerators and heterogeneous clusters.

AB - The increased popularity of accelerators and heterogeneous systems has naturally lead to their adoption in the realm of high performance computing (HPC). As the name implies, HPC applications are always striving to achieve greater performance, typically to perform scientific simulations, financial predictions, etc. The high peak computation rates (GFlops) of accelerators make them particularly attractive to programmers creating these compute intensive HPC codes. Charm++ is one of several parallel programming models used to develop HPC applications. As we will discuss, we believe that some features of the Charm++ programming model make it well suited for programming heterogeneous clusters as well as clusters with accelerators. We also describe the various extensions that have been introduced into the Charm++ programming model to add support for accelerators and heterogeneous clusters.

UR - http://www.scopus.com/inward/record.url?scp=85054448445&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85054448445&partnerID=8YFLogxK

U2 - 10.1201/b10376

DO - 10.1201/b10376

M3 - Chapter

AN - SCOPUS:85054448445

SN - 9781439825365

SP - 393

EP - 412

BT - Scientific Computing with Multicore and Accelerators

PB - CRC Press

ER -