Predictive Modeling in a Polyhedral Optimization Space - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Journal Articles International Journal of Parallel Programming Year : 2013

Predictive Modeling in a Polyhedral Optimization Space


High-level program optimizations, such as loop transformations, are critical for high performance on multi-core targets. However, complex sequences of loop transformations are often required to expose parallelism (both coarse-grain and fine-grain) and improve data locality. The polyhedral compilation framework has proved to be very effective at representing these complex sequences and restructuring compute-intensive applications, seamlessly handling perfectly and imperfectly nested loops. Nevertheless identifying the most effective loop transformations remains a major challenge. We address the problem of selecting the best polyhedral optimizations with dedicated machine learning models, trained specifically on the target machine. We show that these models can quickly select high-performance optimizations with very limited iterative search. Our end-to-end framework is validated using numerous benchmarks on two modern multi-core platforms. We investigate a variety of different machine learning algorithms and hardware counters, and we obtain performance improvements over productions compilers ranging on average from 3.2x to 8.7x, by running not more than 6 program variants from a polyhedral optimization space.
Fichier principal
Vignette du fichier
ijpp-article.13_3.pdf (418.51 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-00918653 , version 1 (13-12-2013)



Eunjung Park, John Cavazos, Louis-Noël Pouchet, Cédric Bastoul, Albert Cohen, et al.. Predictive Modeling in a Polyhedral Optimization Space. International Journal of Parallel Programming, 2013, 41 (5), pp.704--750. ⟨10.1007/s10766-013-0241-1⟩. ⟨hal-00918653⟩
552 View
756 Download



Gmail Facebook X LinkedIn More