Impact of serial scaling of multi-threaded programs in many-core era - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2014

Impact of serial scaling of multi-threaded programs in many-core era

Surya Narayanan
  • Function : Correspondent author
  • PersonId : 949403

Connectez-vous pour contacter l'auteur
André Seznec

Abstract

—Estimating the potential performance of parallel applications on the yet-to-be-designed future many cores is very speculative. The traditional laws used to predict performance of an application do not reflect on the various scaling behaviour of a multi-threaded (MT) application leading to optimistic estimation of performance in manycore era. In this paper, we study the scaling behavior of MT applications as a function of input workload size and the number of cores. For some MT applications in the benchmark suites we analysed, our study shows that the serial fraction in the program increases with input workload size and can be a scalability-limiting factor. Similar to previous studies [5] , we find that using a powerful core (heterogeneous architecture) to execute this serial part of the program can mitigate the impact of serial scaling and improve the overall performance of an application in many-core era.
Fichier principal
Vignette du fichier
WAMCA.pdf (343.14 Ko) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-01089446 , version 1 (01-12-2014)

Identifiers

Cite

Surya Narayanan, Bharath Narasimha Swamy, André Seznec. Impact of serial scaling of multi-threaded programs in many-core era. WAMCA - 5th Workshop on Applications for Multi-Core Architectures, Oct 2014, Paris, France. ⟨10.1109/SBAC-PADW.2014.9⟩. ⟨hal-01089446⟩
466 View
460 Download

Altmetric

Share

Gmail Facebook X LinkedIn More