Compiler-assisted adaptive program scheduling in big.LITTLE systems

Abstract : Energy-aware architectures provide applications with a mix of low and high frequency cores. Selecting the best core configurations for running programs is very challenging. Here, we leverage compilation, runtime monitoring and machine learning to map program phases to their best matching configurations. As a proof-of-concept, we devise the Astro system to show that our approach can outperform a state-of-the-art Linux scheduler for heterogeneous architectures.
Complete list of metadatas
Contributor : Abdoulaye Gamatié <>
Submitted on : Monday, April 15, 2019 - 4:55:41 PM
Last modification on : Saturday, January 18, 2020 - 1:18:23 AM

Links full text




Marcelo Novaes, Vinicius Petrucci, Abdoulaye Gamatié, Fernando Magno Quintão Pereira. Compiler-assisted adaptive program scheduling in big.LITTLE systems. PPoPP: Principles and Practice of Parallel Programming, Feb 2019, Washington, United States. pp.429-430, ⟨10.1145/3293883.3301493⟩. ⟨lirmm-02100287⟩



Record views