Influence of Tasks Duration Variability on Task-Based Runtime Schedulers - Archive ouverte HAL Access content directly
Reports (Research Report) Year : 2018

Influence of Tasks Duration Variability on Task-Based Runtime Schedulers

(1) , (1) , (1)
1

Abstract

In the context of HPC platforms, individual nodes nowadays consist in heterogenous processing resources such as GPU units and multicores. Those resources share communication and storage resources , what induces complex co-scheduling effects, and makes it hard to predict the exact duration of a task or of a communication. To cope with these issues, runtime dynamic schedulers such as StarPU have been developed. These systems base their decisions at runtime on the state of the platform and possibly on static priorities of tasks computed offline. In this paper, our goal is to quantify performance variability in the context of HPC heterogeneous nodes, by focusing on very regular dense linear algebra kernels. Then, we analyze the impact of this variability on a dynamic runtime scheduler such as StarPU, in order to analyze whether the strategies that have been designed in the context of MapReduce applications to cope with stragglers could be transferred to HPC systems, or if the dynamic nature of runtime schedulers is enough to cope with actual performance variations.
Fichier principal
Vignette du fichier
report.pdf (599.08 Ko) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-01716489 , version 1 (23-02-2018)
hal-01716489 , version 2 (28-02-2018)

Identifiers

  • HAL Id : hal-01716489 , version 2

Cite

Olivier Beaumont, Lionel Eyraud-Dubois, Yihong Gao. Influence of Tasks Duration Variability on Task-Based Runtime Schedulers. [Research Report] INRIA. 2018. ⟨hal-01716489v2⟩
207 View
248 Download

Share

Gmail Facebook Twitter LinkedIn More