This thesis aims to explore feasible methods to profile, from the energy efficiency point of view, scientific applications by means of the performance counters and the model-specific registers provided by Intel family microprocessors. The Aurora HPC system used as testbed was made available by courtesy of CO.S.IN.T., and it belongs to the same family of machines that enabled the Eurora cluster hosted at Cineca to reach the 1st rank in the 2013 Green500. Retrieving power consumption information through NVIDIA accelerators and external monitoring and management devices on such platform is explored too.

Application-level energy profiling: the CO.S.IN.T case study

-
2015-12-18

Abstract

This thesis aims to explore feasible methods to profile, from the energy efficiency point of view, scientific applications by means of the performance counters and the model-specific registers provided by Intel family microprocessors. The Aurora HPC system used as testbed was made available by courtesy of CO.S.IN.T., and it belongs to the same family of machines that enabled the Eurora cluster hosted at Cineca to reach the 1st rank in the 2013 Green500. Retrieving power consumption information through NVIDIA accelerators and external monitoring and management devices on such platform is explored too.
18-dic-2015
Baricevic, Moreno
Cozzini, Stefano; Bartolini, Andrea
File in questo prodotto:
File Dimensione Formato  
1963_35155_Baricevic_tesi.pdf

accesso aperto

Tipologia: Tesi
Licenza: Non specificato
Dimensione 21.77 MB
Formato Adobe PDF
21.77 MB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/20.500.11767/5123
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo

Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact