A customized precision format based on mantissa segmentation for accelerating sparse linear algebra
Impacto
Scholar |
Otros documentos de la autoría: Grützmacher, Thomas; Cojean, Terry; Flegar, Goran; Göbel, Fritz; Anzt, Hartwig
Metadatos
Mostrar el registro completo del ítemcomunitat-uji-handle:10234/9
comunitat-uji-handle2:10234/7036
comunitat-uji-handle3:10234/8620
comunitat-uji-handle4:
INVESTIGACIONEste recurso está restringido
https://doi.org/10.1002/cpe.5418 |
Metadatos
Título
A customized precision format based on mantissa segmentation for accelerating sparse linear algebraFecha de publicación
2019Editor
WileyISSN
1532-0626; 1532-0634Cita bibliográfica
Grützmacher T, Cojean T, Flegar G, Göbel F, Anzt H. A customized precision format based on mantissasegmentationforacceleratingsparselinearalgebra.ConcurrencyComputatPractExper.2019;e5418.https://doi.org/10.1002/cpe.5418Tipo de documento
info:eu-repo/semantics/articleVersión de la editorial
https://onlinelibrary.wiley.com/doi/full/10.1002/cpe.5418Versión
info:eu-repo/semantics/publishedVersionPalabras clave / Materias
Resumen
In this work, we pursue the idea of radically decoupling the floating point format used for arithmetic operations from the format used to store the data in memory. We complement this idea with a customized precision ... [+]
In this work, we pursue the idea of radically decoupling the floating point format used for arithmetic operations from the format used to store the data in memory. We complement this idea with a customized precision memory format derived by splitting the mantissa (significand) of standard IEEE formats into segments, such that values can be accessed faster if lower accuracy is acceptable. Combined with precision‐aware algorithms that dynamically adapt the data access accuracy to the numerical requirements, the customized precision memory format can render attractive runtime savings without impacting the memory footprint of the data or the accuracy of the final result. In an experimental analysis using the adaptive precision Jacobi method on diagonalizable test problems, we assess the benefits of the mantissa‐segmenting customized precision format on recent multi‐ and manycore architectures. [-]
Publicado en
Concurrency and Computation: Practice and Experience, 2019Proyecto de investigación
Helmholtz Association. Grant Number: VH-NG-1241; CICYT. Grant Number: TIN2017-82972-R; Swiss National Supercomputing Centre. Grant Number: #d100; Juelich Supercomputing CentreDerechos de acceso
Copyright © John Wiley & Sons, Inc.
http://rightsstatements.org/vocab/InC/1.0/
info:eu-repo/semantics/restrictedAccess
http://rightsstatements.org/vocab/InC/1.0/
info:eu-repo/semantics/restrictedAccess
Aparece en las colecciones
- ICC_Articles [430]