Cargando…

Multi-threaded Object Streaming

The CMS experiment at CERNs Large Hadron Collider in Geneva redesigned the code handling the conditions data during the last years, aiming to increase performance and enhance maintainability. The new design includes a move to serialise all payloads before storing them into the database, allowing the...

Descripción completa

Detalles Bibliográficos
Autores principales: Pfeiffer, Andreas, Di Guida, Salvatore, Govi, Giacomo, Ojeda, Miguel, Sipos, Roland
Lenguaje:eng
Publicado: 2015
Materias:
Acceso en línea:https://dx.doi.org/10.1088/1742-6596/664/4/042044
http://cds.cern.ch/record/2121258
_version_ 1780949352230944768
author Pfeiffer, Andreas
Di Guida, Salvatore
Govi, Giacomo
Ojeda, Miguel
Sipos, Roland
author_facet Pfeiffer, Andreas
Di Guida, Salvatore
Govi, Giacomo
Ojeda, Miguel
Sipos, Roland
author_sort Pfeiffer, Andreas
collection CERN
description The CMS experiment at CERNs Large Hadron Collider in Geneva redesigned the code handling the conditions data during the last years, aiming to increase performance and enhance maintainability. The new design includes a move to serialise all payloads before storing them into the database, allowing the handling of the payloads in external tools independent of a given software release. In this talk we present the results of performance studies done using the serialisation package from the Boost suite as well as serialisation done with the ROOT (v5) tools. Furthermore, as the Boost tools allow parallel (de-)serialisation, we show the performance gains achieved with parallel threads when de-serialising a realistic set of conditions in CMS. Without specific optimisations an overall speed up of a factor of 3-4 was achieved using multi-threaded loading and de-serialisation of our conditions.
id cern-2121258
institution Organización Europea para la Investigación Nuclear
language eng
publishDate 2015
record_format invenio
spelling cern-21212582022-08-10T13:07:48Zdoi:10.1088/1742-6596/664/4/042044http://cds.cern.ch/record/2121258engPfeiffer, AndreasDi Guida, SalvatoreGovi, GiacomoOjeda, MiguelSipos, RolandMulti-threaded Object StreamingDetectors and Experimental TechniquesThe CMS experiment at CERNs Large Hadron Collider in Geneva redesigned the code handling the conditions data during the last years, aiming to increase performance and enhance maintainability. The new design includes a move to serialise all payloads before storing them into the database, allowing the handling of the payloads in external tools independent of a given software release. In this talk we present the results of performance studies done using the serialisation package from the Boost suite as well as serialisation done with the ROOT (v5) tools. Furthermore, as the Boost tools allow parallel (de-)serialisation, we show the performance gains achieved with parallel threads when de-serialising a realistic set of conditions in CMS. Without specific optimisations an overall speed up of a factor of 3-4 was achieved using multi-threaded loading and de-serialisation of our conditions.CMS-CR-2015-057oai:cds.cern.ch:21212582015-05-11
spellingShingle Detectors and Experimental Techniques
Pfeiffer, Andreas
Di Guida, Salvatore
Govi, Giacomo
Ojeda, Miguel
Sipos, Roland
Multi-threaded Object Streaming
title Multi-threaded Object Streaming
title_full Multi-threaded Object Streaming
title_fullStr Multi-threaded Object Streaming
title_full_unstemmed Multi-threaded Object Streaming
title_short Multi-threaded Object Streaming
title_sort multi-threaded object streaming
topic Detectors and Experimental Techniques
url https://dx.doi.org/10.1088/1742-6596/664/4/042044
http://cds.cern.ch/record/2121258
work_keys_str_mv AT pfeifferandreas multithreadedobjectstreaming
AT diguidasalvatore multithreadedobjectstreaming
AT govigiacomo multithreadedobjectstreaming
AT ojedamiguel multithreadedobjectstreaming
AT siposroland multithreadedobjectstreaming