Open AccessSplit Tiling for GPUs: Automatic Parallelization Using Trapezoidal Tiles to Reconcile Parallelism and Locality, avoiding Divergence and Load Imbalance
Author(s) -
Albert Cohen,
Tobias Grosser,
Paul H. J. Kelly,
J. Ramanujam,
P. Sadayappan,
Sven Verdoolaege
Publication year - 2013
Publication title -
hal (le centre pour la communication scientifique directe)
Language(s) - English
Resource type - Conference proceedings
Subject(s) - loop tiling , parallel computing , computer science , stencil , nested loop join , compiler , locality , cuda , algorithm , computation , computational science , linguistics , philosophy , programming language