Energy consumption represents one of the most relevant issues by now in operating HPC systems for scientific applications. The investigation and assessment of unconventional processors with high ratio of performance per watt is interesting for a better trade-off between time-to-solution and energy-to-solution. Computing on SoC Architectures (COSA) is a threeyear project (2015-2017) funded by the Scientific Commission V of INFN and led by CNAF, which aims to investigate the performances and the total cost of ownership offered by computing systems based on commodity low power Systems on Chip (SoC) and high energy-efficient systems based on GP-GPUs. Here we present the results of the project and we describe the methodology we have used to measure energy performance and the tools we have implemented to monitor the power drained by applications while running.
COmputing on SoC Architectures: the INFN COSA project / Cesini, D.; Corni, E.; Falabella, A.; Ferraro, A.; Guidi, G.; Morganti, L.; Calore, E.; Schifano, S. F.; Michelotto, M.; Alfieri, R.; De Pietri, R.; Boccali, T.; Biagioni, A.; Lo Cicero, F.; Lonardo, A.; Martinelli, M.; Paolucci, P. S.; Pastorelli, E.; Vicini, P.. - In: INFN-CNAF ANNUAL REPORT. - ISSN 2283-5490. - (2017), pp. 231-236.
COmputing on SoC Architectures: the INFN COSA project
R. Alfieri
;R. De Pietri
;
2017-01-01
Abstract
Energy consumption represents one of the most relevant issues by now in operating HPC systems for scientific applications. The investigation and assessment of unconventional processors with high ratio of performance per watt is interesting for a better trade-off between time-to-solution and energy-to-solution. Computing on SoC Architectures (COSA) is a threeyear project (2015-2017) funded by the Scientific Commission V of INFN and led by CNAF, which aims to investigate the performances and the total cost of ownership offered by computing systems based on commodity low power Systems on Chip (SoC) and high energy-efficient systems based on GP-GPUs. Here we present the results of the project and we describe the methodology we have used to measure energy performance and the tools we have implemented to monitor the power drained by applications while running.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
