Recent developments in Big Data frameworks are moving towards reservation based approaches as a mean to manage the increasingly complex mix of computations, whereas preemption techniques are employed to meet strict jobs deadlines. Within this work we propose and evaluate a new planning algorithm in the context of reservation based scheduling. Our approach is able to achieve high cluster utilization while minimizing the need for preemption that causes system overheads and planning mispredictions.
Preemption-aware planning on Big-Data systems / M. Rabozzi, M. Mazzucchelli, R. Cordone, G.M. Fumarola, M.D. Santambrogio - In: Proceedings of the ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming, PPOPP[s.l] : Association for Computing Machinery, 2016. - ISBN 9781450340922. - pp. 1-2 (( Intervento presentato al 2. convegno Principles and Practice of Parallel Programming tenutosi a Barcelona nel 2016 [10.1145/2851141.2851187].
Preemption-aware planning on Big-Data systems
R. Cordone;
2016
Abstract
Recent developments in Big Data frameworks are moving towards reservation based approaches as a mean to manage the increasingly complex mix of computations, whereas preemption techniques are employed to meet strict jobs deadlines. Within this work we propose and evaluate a new planning algorithm in the context of reservation based scheduling. Our approach is able to achieve high cluster utilization while minimizing the need for preemption that causes system overheads and planning mispredictions.File | Dimensione | Formato | |
---|---|---|---|
paper.pdf
accesso riservato
Descrizione: Conferenza
Tipologia:
Post-print, accepted manuscript ecc. (versione accettata dall'editore)
Dimensione
214.68 kB
Formato
Adobe PDF
|
214.68 kB | Adobe PDF | Visualizza/Apri Richiedi una copia |
Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.