Articles | Volume 9, issue 8
Geosci. Model Dev., 9, 2881–2892, 2016
https://doi.org/10.5194/gmd-9-2881-2016
Geosci. Model Dev., 9, 2881–2892, 2016
https://doi.org/10.5194/gmd-9-2881-2016

Development and technical paper 26 Aug 2016

Development and technical paper | 26 Aug 2016

Asynchronous communication in spectral-element and discontinuous Galerkin methods for atmospheric dynamics – a case study using the High-Order Methods Modeling Environment (HOMME-homme_dg_branch)

Benjamin F. Jamroz1 and Robert Klöfkorn1,2 Benjamin F. Jamroz and Robert Klöfkorn
  • 1Computational Information Systems Laboratory, National Center for Atmospheric Research, 1850 Table Mesa Drive, Boulder, CO 80305, USA
  • 2International Research Institute of Stavanger, P.O. Box 8046, 4068 Stavanger, Norway

Abstract. The scalability of computational applications on current and next-generation supercomputers is increasingly limited by the cost of inter-process communication. We implement non-blocking asynchronous communication in the High-Order Methods Modeling Environment for the time integration of the hydrostatic fluid equations using both the spectral-element and discontinuous Galerkin methods. This allows the overlap of computation with communication, effectively hiding some of the costs of communication. A novel detail about our approach is that it provides some data movement to be performed during the asynchronous communication even in the absence of other computations. This method produces significant performance and scalability gains in large-scale simulations.

Download
Short summary
The scalability of computational applications on current and next-generation supercomputers is increasingly limited by the cost of inter-process communication. We implement communication hiding data exchange in the High-Order Methods Modeling Environment (HOMME) for the time integration of the hydrostatic fluid equations using both the spectral-element and discontinuous Galerkin methods. The presented approach produces significant performance and scalability gains in large-scale simulations.