Fastsg: A fast routines library for sparse grids

Alin Muraraşu; Gerrit Buse; Dirk Pflüger; Josef Weidendorfer; Arndt Bode

doi:10.1016/j.procs.2012.04.038

Fastsg: A fast routines library for sparse grids

Alin Muraraşu, Gerrit Buse, Dirk Pflüger, Josef Weidendorfer, Arndt Bode

Informatics 10 - Chair of Computer Architecture and Parallel Systems

Technical University of Munich

Research output: Contribution to journal › Conference article › peer-review

14 Scopus citations

Abstract

In a complex processor landscape dominated by multi- and many-core processors, simplifying programming plays a crucial role in enhancing developers' productivity. One way is to use highly tuned library functions. In this paper we present fastsg, an optimized library for the sparse grid technique with support for dimensional truncation. With optimizations for best cache use and vectorization, we improve the performance on one processor core up to a factor of 10. Parallelization using OpenMP scales almost linearly on a 12-core system.

Original language	English
Pages (from-to)	354-363
Number of pages	10
Journal	Procedia Computer Science
Volume	9
DOIs	https://doi.org/10.1016/j.procs.2012.04.038
State	Published - 2012
Event	12th Annual International Conference on Computational Science, ICCS 2012 - Omaha, NB, United States Duration: 4 Jun 2012 → 6 Jun 2012

Keywords

Dimensional truncation
Library
Optimizations
Sparse grids

Access to Document

10.1016/j.procs.2012.04.038

Cite this

@article{138d6fa9a71f45d7906f599351774ed2,

title = "Fastsg: A fast routines library for sparse grids",

abstract = "In a complex processor landscape dominated by multi- and many-core processors, simplifying programming plays a crucial role in enhancing developers' productivity. One way is to use highly tuned library functions. In this paper we present fastsg, an optimized library for the sparse grid technique with support for dimensional truncation. With optimizations for best cache use and vectorization, we improve the performance on one processor core up to a factor of 10. Parallelization using OpenMP scales almost linearly on a 12-core system.",

keywords = "Dimensional truncation, Library, Optimizations, Sparse grids",

author = "Alin Murara{\c s}u and Gerrit Buse and Dirk Pfl{\"u}ger and Josef Weidendorfer and Arndt Bode",

year = "2012",

doi = "10.1016/j.procs.2012.04.038",

language = "English",

volume = "9",

pages = "354--363",

journal = "Procedia Computer Science",

issn = "1877-0509",

publisher = "Elsevier BV",

note = "12th Annual International Conference on Computational Science, ICCS 2012 ; Conference date: 04-06-2012 Through 06-06-2012",

}

TY - JOUR

T1 - Fastsg

T2 - 12th Annual International Conference on Computational Science, ICCS 2012

AU - Muraraşu, Alin

AU - Buse, Gerrit

AU - Pflüger, Dirk

AU - Weidendorfer, Josef

AU - Bode, Arndt

PY - 2012

Y1 - 2012

N2 - In a complex processor landscape dominated by multi- and many-core processors, simplifying programming plays a crucial role in enhancing developers' productivity. One way is to use highly tuned library functions. In this paper we present fastsg, an optimized library for the sparse grid technique with support for dimensional truncation. With optimizations for best cache use and vectorization, we improve the performance on one processor core up to a factor of 10. Parallelization using OpenMP scales almost linearly on a 12-core system.

AB - In a complex processor landscape dominated by multi- and many-core processors, simplifying programming plays a crucial role in enhancing developers' productivity. One way is to use highly tuned library functions. In this paper we present fastsg, an optimized library for the sparse grid technique with support for dimensional truncation. With optimizations for best cache use and vectorization, we improve the performance on one processor core up to a factor of 10. Parallelization using OpenMP scales almost linearly on a 12-core system.

KW - Dimensional truncation

KW - Library

KW - Optimizations

KW - Sparse grids

UR - http://www.scopus.com/inward/record.url?scp=84870736957&partnerID=8YFLogxK

U2 - 10.1016/j.procs.2012.04.038

DO - 10.1016/j.procs.2012.04.038

M3 - Conference article

AN - SCOPUS:84870736957

SN - 1877-0509

VL - 9

SP - 354

EP - 363

JO - Procedia Computer Science

JF - Procedia Computer Science

Y2 - 4 June 2012 through 6 June 2012

ER -

Fastsg: A fast routines library for sparse grids

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this