A MPI back-end for the OpenACC accULL. Exploiting OpenACC semantics in Message Passing Clusters
Dept. de E. I. O. y Computacion Universidad de La Laguna, 38271 La Laguna, Spain
EPCC, HPC-Europa2 Report, 2012
@article{fumero2012mpi,
title={A MPI back-end for the OpenACC accull. Exploiting OpenACC semantics in Message Passing Clusters},
author={Fumero, J.J.},
year={2012}
}
The irruption in the HPC scene of hardware acceletarors has made available unprecedented performance to developers. However, even expert developers may not be ready to exploit the complex hierarchies of these new heterogeneous systems. We need to find a way to leverage the programming effort in these architectures at programming language level, otherwise, developers will spend most of their time focusing on device-specific code instead of implementing algorithmic enhancements. In this work we present an extension of our compiler novel OpenACC implementation accULL to execute MPI over OpenCL programs in a cluster platform.
December 4, 2012 by hgpu
