A GPU offloading mechanism for LHCb

The current computational infrastructure at LHCb is designed for sequential execution. It is possible to make use of modern multi-core machines by using multi-threaded algorithms and running multiple instances in parallel, but there is no way to make efficient use of specialized massively parallel h...

Descripción completa

Detalles Bibliográficos
Autores principales: Badalov, Alexey, Campora Perez, Daniel Hugo, Zvyagin, Alexander, Neufeld, Niko, Vilasis Cardona, Xavier
Lenguaje:eng
Publicado: 2014
Materias:
Computing and Computers
Acceso en línea:https://dx.doi.org/10.1088/1742-6596/513/5/052004
http://cds.cern.ch/record/2026341
_version_ 1780947351492362240
author Badalov, Alexey
Campora Perez, Daniel Hugo
Zvyagin, Alexander
Neufeld, Niko
Vilasis Cardona, Xavier
author_facet Badalov, Alexey
Campora Perez, Daniel Hugo
Zvyagin, Alexander
Neufeld, Niko
Vilasis Cardona, Xavier
author_sort Badalov, Alexey
collection CERN
description The current computational infrastructure at LHCb is designed for sequential execution. It is possible to make use of modern multi-core machines by using multi-threaded algorithms and running multiple instances in parallel, but there is no way to make efficient use of specialized massively parallel hardware, such as graphical processing units and Intel Xeon/Phi. We extend the current infrastructure with an out-of-process computational server able to gather data from multiple instances and process them in large batches.
id oai-inspirehep.net-1302130
institution Organización Europea para la Investigación Nuclear
language eng
publishDate 2014
record_format invenio
spelling oai-inspirehep.net-13021302022-08-17T13:29:09Zdoi:10.1088/1742-6596/513/5/052004http://cds.cern.ch/record/2026341engBadalov, AlexeyCampora Perez, Daniel HugoZvyagin, AlexanderNeufeld, NikoVilasis Cardona, XavierA GPU offloading mechanism for LHCbComputing and ComputersThe current computational infrastructure at LHCb is designed for sequential execution. It is possible to make use of modern multi-core machines by using multi-threaded algorithms and running multiple instances in parallel, but there is no way to make efficient use of specialized massively parallel hardware, such as graphical processing units and Intel Xeon/Phi. We extend the current infrastructure with an out-of-process computational server able to gather data from multiple instances and process them in large batches.oai:inspirehep.net:13021302014
spellingShingle Computing and Computers
Badalov, Alexey
Campora Perez, Daniel Hugo
Zvyagin, Alexander
Neufeld, Niko
Vilasis Cardona, Xavier
A GPU offloading mechanism for LHCb
title A GPU offloading mechanism for LHCb
title_full A GPU offloading mechanism for LHCb
title_fullStr A GPU offloading mechanism for LHCb
title_full_unstemmed A GPU offloading mechanism for LHCb
title_short A GPU offloading mechanism for LHCb
title_sort gpu offloading mechanism for lhcb
topic Computing and Computers
url https://dx.doi.org/10.1088/1742-6596/513/5/052004
http://cds.cern.ch/record/2026341
work_keys_str_mv AT badalovalexey agpuoffloadingmechanismforlhcb
AT camporaperezdanielhugo agpuoffloadingmechanismforlhcb
AT zvyaginalexander agpuoffloadingmechanismforlhcb
AT neufeldniko agpuoffloadingmechanismforlhcb
AT vilasiscardonaxavier agpuoffloadingmechanismforlhcb
AT badalovalexey gpuoffloadingmechanismforlhcb
AT camporaperezdanielhugo gpuoffloadingmechanismforlhcb
AT zvyaginalexander gpuoffloadingmechanismforlhcb
AT neufeldniko gpuoffloadingmechanismforlhcb
AT vilasiscardonaxavier gpuoffloadingmechanismforlhcb

Ejemplares similares