Proceedings chapter
OA Policy
English

Efficient scheduling policies for dynamic data flow programs executed on multi-core

Presented atLausanne (Suisse), 7-9 septembre
Publication date2016
Abstract

An important challenge of dataflow program implementations on multi-core platforms is the partitioning and scheduling providing the best possible throughput when satisfying multiple objective functions. Not only it has been proven that these problems are NP-complete, but also the quality of any heuristic approach can be affected by other factors (e.g., buffer dimensioning, influence of an established partitioning configuration and scheduling strategy on each other, uncertainties of a compiler affecting the profiling information). This paper proposes an evaluation of alternative partitioning and scheduling configurations based on the application profiling results. It investigates the impact of the scheduling on the overall execution time, and verifies which policies could further drive the metaheuristic-based search of a close-to-optimal partitioning configuration.

Citation (ISO format)
MICHALSKA, Malgorzata et al. Efficient scheduling policies for dynamic data flow programs executed on multi-core. In: 11th International Meeting on Logistics Research. Lausanne (Suisse). [s.l.] : [s.n.], 2016.
Main files (1)
Proceedings chapter (Accepted version)
accessLevelPublic
Identifiers
  • PID : unige:91453
504views
180downloads

Technical informations

Creation30/01/2017 09:13:00
First validation30/01/2017 09:13:00
Update time15/03/2023 01:19:47
Status update15/03/2023 01:19:47
Last indexation31/10/2024 06:00:54
All rights reserved by Archive ouverte UNIGE and the University of GenevaunigeBlack