Adaptive Partitioning for Iterated Sequences of Irregular OpenCL Kernels

Pierre Huchant, Denis Barthou, Marie-Christine Counilh
Inria / LaBRI, Bordeaux INP, U. of Bordeaux, Talence, France
30th International Symposium on Computer Architecture and High Performance Computing, 2018


   title={Adaptive Partitioning for Iterated Sequences of Irregular OpenCL Kernels},

   author={Huchant, Pierre and Barthou, Denis and Counilh, Marie-Christine},




Download Download (PDF)   View View   Source Source   



OpenCL defines a common parallel programming language for all devices, although writing tasks adapted to the devices, managing communication and load-balancing issues are left to the programmer. We propose in this paper a static/dynamic approach for the execution of an iterated sequence of data-dependent kernels on a multi-device heterogeneous architecture. The method allows to automatically distribute irregular kernels onto multiple devices and tackles, without training, both load balancing and data transfers issues coming from hardware heterogeneity, load imbalance within the application itself and load variations between repeated executions of the sequence.
No votes yet.
Please wait...

* * *

* * *

HGPU group © 2010-2021 hgpu.org

All rights belong to the respective authors

Contact us: