The constantly increasing gap between communication and computation performance emphasizes the importance of communication-avoidance techniques. Caching is a well-known concept used to reduce accesses to slow local memories. In this work, we extend the caching idea to MPI3 Remote Memory Access (RMA) operations. Here, caching can avoid inter-node communications and achieve similar benefits for irregular applications as communication-avoiding algorithms for structured applications. We propose CLaMPI, a caching library layered on top of MPI-3 RMA, to automatically optimize code with minimum user intervention. We demonstrate how cached RMA improves the performance of a Barnes Hut simulation and a Local Clustering Coefficient computation up to a factor of 1.8x and 5x, respectively. Due to the low overheads in the cache miss case and the potential benefits, we expect that our ideas around transparent RMA caching will soon be an integral part of many MPI libraries

Transparent caching for RMA systems / Girolamo, Salvatore DI; Vella, Flavio; Hoefler, Torsten. - (2017), pp. 1018-1027. (Intervento presentato al convegno 31st IEEE International Parallel and Distributed Processing Symposium, IPDPS 2017 tenutosi a Usa nel 2017) [10.1109/IPDPS.2017.92].

Transparent caching for RMA systems

Vella, Flavio;
2017

Abstract

The constantly increasing gap between communication and computation performance emphasizes the importance of communication-avoidance techniques. Caching is a well-known concept used to reduce accesses to slow local memories. In this work, we extend the caching idea to MPI3 Remote Memory Access (RMA) operations. Here, caching can avoid inter-node communications and achieve similar benefits for irregular applications as communication-avoiding algorithms for structured applications. We propose CLaMPI, a caching library layered on top of MPI-3 RMA, to automatically optimize code with minimum user intervention. We demonstrate how cached RMA improves the performance of a Barnes Hut simulation and a Local Clustering Coefficient computation up to a factor of 1.8x and 5x, respectively. Due to the low overheads in the cache miss case and the potential benefits, we expect that our ideas around transparent RMA caching will soon be an integral part of many MPI libraries
2017
31st IEEE International Parallel and Distributed Processing Symposium, IPDPS 2017
Barnes hut; caching; irregular applications; local clustering coefficient; MPI; nbody; remote transfer; RMA; Information Systems; Computer Networks and Communications; Hardware and Architecture
04 Pubblicazione in atti di convegno::04b Atto di convegno in volume
Transparent caching for RMA systems / Girolamo, Salvatore DI; Vella, Flavio; Hoefler, Torsten. - (2017), pp. 1018-1027. (Intervento presentato al convegno 31st IEEE International Parallel and Distributed Processing Symposium, IPDPS 2017 tenutosi a Usa nel 2017) [10.1109/IPDPS.2017.92].
File allegati a questo prodotto
File Dimensione Formato  
Vella_RMA-systems.pdf

solo gestori archivio

Tipologia: Versione editoriale (versione pubblicata con il layout dell'editore)
Licenza: Tutti i diritti riservati (All rights reserved)
Dimensione 409.74 kB
Formato Adobe PDF
409.74 kB Adobe PDF   Contatta l'autore

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/11573/1068895
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 5
  • ???jsp.display-item.citation.isi??? 3
social impact