A Compiler Algorithm to Guide Runtime Scheduling

Christophe Alias 1 Samuel Thibault 2, 3 Laure Gonnord 1
1 CASH - CASH - Compilation and Analysis, Software and Hardware
Inria Grenoble - Rhône-Alpes, LIP - Laboratoire de l'Informatique du Parallélisme
3 STORM - STatic Optimizations, Runtime Methods
LaBRI - Laboratoire Bordelais de Recherche en Informatique, Inria Bordeaux - Sud-Ouest
Abstract : Task-level parallelism is usually exploited by a runtime scheduler, after tasks are mapped to processing units by a compiler. In this report, we propose a compilation-centric runtime scheduling strategy. We propose a complete compilation algorithm to split the tasks in three parts, whose properties are intended to help the scheduler to take the right decisions. In particular, we show how the polyhedral model may provide a precious help to compute tricky scheduling and parallelism informations. Our compiler is available and may be tried online at http://foobar. ens-lyon.fr/kut.
Complete list of metadatas

Cited literature [8 references]  Display  Hide  Download

https://hal.inria.fr/hal-02421327
Contributor : Christophe Alias <>
Submitted on : Friday, December 20, 2019 - 1:41:23 PM
Last modification on : Wednesday, January 8, 2020 - 1:45:01 AM

File

RR-9315.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-02421327, version 1

Citation

Christophe Alias, Samuel Thibault, Laure Gonnord. A Compiler Algorithm to Guide Runtime Scheduling. [Research Report] RR-9315, INRIA Grenoble; INRIA Bordeaux - Sud-Ouest. 2019. ⟨hal-02421327⟩

Share

Metrics

Record views

51

Files downloads

316