High-Performance Computing (HPC) is rapidly moving towards the adoption of nodes characterized by an heterogeneous set of processing resources. This has already shown benets in terms of both performance and energy eciency. On the other side, heterogeneous systems are challenging from the application development and the resource management perspective. In this work, we discuss some outcomes of the MANGO project, showing the results of the execution of real applications on a emulated deeply heterogeneous systems for HPC. Moreover, we assessed the achievements of a proposed resource allocation policy, aiming at identifying a priori the best resource allocation options for a starting application.

Predictive Resource Management for Next-generation High-Performance Computing Heterogeneous Platforms

Giuseppe Massari;Anna Pupykina;Giovanni Agosta;William Fornaciari
2019

Abstract

High-Performance Computing (HPC) is rapidly moving towards the adoption of nodes characterized by an heterogeneous set of processing resources. This has already shown benets in terms of both performance and energy eciency. On the other side, heterogeneous systems are challenging from the application development and the resource management perspective. In this work, we discuss some outcomes of the MANGO project, showing the results of the execution of real applications on a emulated deeply heterogeneous systems for HPC. Moreover, we assessed the achievements of a proposed resource allocation policy, aiming at identifying a priori the best resource allocation options for a starting application.
Embedded Computer Systems: Architectures, Modeling, and Simulation
978-3-030-27562-4
978-3-030-27561-7
HPC, Heterogeneous Systems, Resource Management
File in questo prodotto:
File Dimensione Formato  
SAMOS_2019_RTRM_MANGO.pdf

accesso aperto

Descrizione: camera ready
: Post-Print (DRAFT o Author’s Accepted Manuscript-AAM)
Dimensione 457.88 kB
Formato Adobe PDF
457.88 kB Adobe PDF Visualizza/Apri
MANGO_SAMOS_2019.pdf

Accesso riservato

Descrizione: versione pubblicata
: Publisher’s version
Dimensione 607.34 kB
Formato Adobe PDF
607.34 kB 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/11311/1099257
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 5
  • ???jsp.display-item.citation.isi??? 5
social impact