The ANTAREX Approach to Autotuning and Adaptivity for Energy Efficient HPC Systems
Cristina Silvano, Giovanni Agosta, Stefano Cherubin, Davide Gadioli, Gianluca Palermo, Andrea Bartolini, Luca Benini, Jan Martinovic, Martin Palkovic, Katerina Slaninova, Joao Bispo, Joao M. P. Cardoso, Rui Abreu, Pedro Pinto, Carlo Cavazzoni, Nico Sanna, Andrea R. Beccari, Radim Cmar, Erven Rohou
title={The ANTAREX Approach to Autotuning and Adaptivity for Energy Efficient HPC Systems},
author={Silvano, Cristina and Agosta, Giovanni and Cherubin, Stefano and Gadioli, Davide and Palermo, Gianluca and Bartolini, Andrea and Benini, Luca and Martinovi{v c}, Jan and Palkovi{v c}, Martin and Slaninov{‘a}, Kate{v r}ina and Bispo, Joao and Cardoso, Joao M. P. and Abreu, Rui and Pinto, Pedro and Cavazzoni, Carlo and Sanna, Nico and Beccari, Andrea R. and Cmar, Radim and Rohou, Erven},
url={https://hal.inria.fr/hal-01341826},
booktitle={ACM International Conference on Computing Frontiers 2016},
address={Como, Italy},
year={2016},
month={May},
doi={10.1145/2903150.2903470},
keywords={High Performance Computing; Autotuning; Adaptivity; DSL; Compilers; Energy Efficiency},
The ANTAREX project aims at expressing the application self-adaptivity through a Domain Specific Language (DSL) and to run-time manage and autotune applications for green and heterogeneous High Performance Computing (HPC) systems up to Exascale. The DSL approach allows the definition of energy-efficiency, performance, and adaptivity strategies as well as their enforcement at runtime through application autotuning and resource and power management. We show through a mini-app extracted from one of the project application use cases some initial exploration of application precision tuning by means enabled by the DSL.
