Data intensive ATLAS workflows in the Cloud

Mostrar otras versiones (3)

This contribution reports on the feasibility of executing data intensive workflows on Cloud infrastructures. In order to assess this, the metric ETC = Events/Time/Cost is formed, which quantifies the different workflow and infrastructure configurations that are tested against each other. In these te...

Descripción completa

Detalles Bibliográficos
Autores principales: Rzehorz, Gerhard Ferdinand, Keeble, Oliver, Kawamura, Gen, Quadt, Arnulf
Lenguaje:eng
Publicado: 2016
Materias:
Particle Physics - Experiment
Acceso en línea:http://cds.cern.ch/record/2221182
_version_ 1780952233323528192
author Rzehorz, Gerhard Ferdinand
Keeble, Oliver
Kawamura, Gen
Quadt, Arnulf
author_facet Rzehorz, Gerhard Ferdinand
Keeble, Oliver
Kawamura, Gen
Quadt, Arnulf
author_sort Rzehorz, Gerhard Ferdinand
collection CERN
description This contribution reports on the feasibility of executing data intensive workflows on Cloud infrastructures. In order to assess this, the metric ETC = Events/Time/Cost is formed, which quantifies the different workflow and infrastructure configurations that are tested against each other. In these tests ATLAS reconstruction Jobs are run, examining the effects of overcommitting (more parallel processes running than CPU cores available), scheduling (staggered execution) and scaling (number of cores). The desirability of commissioning storage in the cloud is evaluated, in conjunction with a simple analytical model of the system, and correlated with questions about the network bandwidth, caches and what kind of storage to utilise. In the end a cost/benefit evaluation of different infrastructure configurations and workflows is undertaken, with the goal to find the maximum of the ETC value
id cern-2221182
institution Organización Europea para la Investigación Nuclear
language eng
publishDate 2016
record_format invenio
spelling cern-22211822019-09-30T06:29:59Zhttp://cds.cern.ch/record/2221182engRzehorz, Gerhard FerdinandKeeble, OliverKawamura, GenQuadt, ArnulfData intensive ATLAS workflows in the CloudParticle Physics - ExperimentThis contribution reports on the feasibility of executing data intensive workflows on Cloud infrastructures. In order to assess this, the metric ETC = Events/Time/Cost is formed, which quantifies the different workflow and infrastructure configurations that are tested against each other. In these tests ATLAS reconstruction Jobs are run, examining the effects of overcommitting (more parallel processes running than CPU cores available), scheduling (staggered execution) and scaling (number of cores). The desirability of commissioning storage in the cloud is evaluated, in conjunction with a simple analytical model of the system, and correlated with questions about the network bandwidth, caches and what kind of storage to utilise. In the end a cost/benefit evaluation of different infrastructure configurations and workflows is undertaken, with the goal to find the maximum of the ETC valueATL-SOFT-SLIDE-2016-752oai:cds.cern.ch:22211822016-10-03
spellingShingle Particle Physics - Experiment
Rzehorz, Gerhard Ferdinand
Keeble, Oliver
Kawamura, Gen
Quadt, Arnulf
Data intensive ATLAS workflows in the Cloud
title Data intensive ATLAS workflows in the Cloud
title_full Data intensive ATLAS workflows in the Cloud
title_fullStr Data intensive ATLAS workflows in the Cloud
title_full_unstemmed Data intensive ATLAS workflows in the Cloud
title_short Data intensive ATLAS workflows in the Cloud
title_sort data intensive atlas workflows in the cloud
topic Particle Physics - Experiment
url http://cds.cern.ch/record/2221182
work_keys_str_mv AT rzehorzgerhardferdinand dataintensiveatlasworkflowsinthecloud
AT keebleoliver dataintensiveatlasworkflowsinthecloud
AT kawamuragen dataintensiveatlasworkflowsinthecloud
AT quadtarnulf dataintensiveatlasworkflowsinthecloud

Ejemplares similares