Towards an exascale enabled sparse solver repository

None, None; None, None; None, None; None, None; None, None; None, None; None, None; None, None; None, None; None, None; None, None; None, None

Towards an exascale enabled sparse solver repository

Conference Paper (2016)

Author(s)

Jonas Thies (Deutsches Zentrum für Luft- und Raumfahrt (DLR))

Martin Galgon (Bergische Universität Wuppertal )

Faisal Shahzad (Friedrich-Alexander-Universität Erlangen-Nürnberg)

Andreas Alvermann (Greifswald University)

Moritz Kreutzer (Friedrich-Alexander-Universität Erlangen-Nürnberg)

Andreas Pieper (Greifswald University)

Melven Röhrig-Zöllner (Deutsches Zentrum für Luft- und Raumfahrt (DLR))

Achim Basermann (Deutsches Zentrum für Luft- und Raumfahrt (DLR))

Holger Fehske (Greifswald University)

Georg Hager (Friedrich-Alexander-Universität Erlangen-Nürnberg)

Bruno Lang (Bergische Universität Wuppertal )

Gerhard Wellein (Friedrich-Alexander-Universität Erlangen-Nürnberg)

DOI related publication

https://doi.org/10.1007/978-3-319-40528-5_13 Final published version

To reference this document use

https://resolver.tudelft.nl/uuid:8195f0e3-459f-42cb-837d-e028153c8b28

More Info

expand_more

Publication Year

2016

Language

English

Pages (from-to)

295-316

Publisher

Springer

ISBN (print)

9783319405261

Event

Downloads counter

207

Abstract

As we approach the exascale computing era, disruptive changes in the software landscape are required to tackle the challenges posed by manycore CPUs and accelerators. We discuss the development of a new ‘exascale enabled’ sparse solver repository (the ESSR) that addresses these challenges—from fundamental design considerations and development processes to actual implementations of some prototypical iterative schemes for computing eigenvalues of sparse matrices. Key features of the ESSR include holistic performance engineering, tight integration between software layers and mechanisms to mitigate hardware failures.