Highly scalable dynamic load balancing in the atmospheric modeling system COSMO-SPECS+FD4

Authors:
Matthias Lieber;Verena Grützun;Ralf Wolke;Matthias S. Müller;Wolfgang E. Nagel
Affiliations:
Center for Information Services and High Performance Computing, TU Dresden, Dresden, Germany;Max Planck Institute for Meteorology, Hamburg, Germany;Leibniz Institute for Tropospheric Research, Leipzig, Germany;Center for Information Services and High Performance Computing, TU Dresden, Dresden, Germany;Center for Information Services and High Performance Computing, TU Dresden, Dresden, Germany
Venue:
PARA'10 Proceedings of the 10th international conference on Applied Parallel and Scientific Computing - Volume Part I
Year:
2010

Citing 6
Cited 1

MM90: a scalable parallel implementation of the Penn State/NCAR Mesoscale Model (MM5)

Parallel Computing - Special issue on applications: parallel computing in regional weather modeling
Zoltan Data Management Service for Parallel Dynamic Applications

Computing in Science and Engineering
Fast optimal load balancing algorithms for 1D partitioning

Journal of Parallel and Distributed Computing
The Model Coupling Toolkit: A New Fortran90 Toolkit for Building Multiphysics Parallel Coupled Models

International Journal of High Performance Computing Applications
Optimizing the coupling in parallel air quality model systems

Environmental Modelling & Software
Scalability challenges for massively parallel AMR applications

IPDPS '09 Proceedings of the 2009 IEEE International Symposium on Parallel&Distributed Processing

AMRC: an algebraic model for reconfiguration of high performance cluster computing systems at runtime

The Journal of Supercomputing

Quantified Score

Hi-index	0.00

Visualization

Abstract

To study the complex interactions between cloud processes and the atmosphere, several atmospheric models have been coupled with detailed spectral cloud microphysics schemes. These schemes are computationally expensive, which limits their practical application. Additionally, our performance analysis of the model system COSMO-SPECS (atmospheric model of the Consortium for Small-scale Modeling coupled with SPECtral bin cloud microphysicS) shows a significant load imbalance due to the cloud model. To overcome this issue and enable dynamic load balancing, we propose the separation of the cloud scheme from the static partitioning of the atmospheric model. Using the framework FD4 (Four-Dimensional Distributed Dynamic Data structures), we show that this approach successfully eliminates the load imbalance and improves the scalability of the model system. We present a scalability analysis of the dynamic load balancing and coupling for two different supercomputers. The observed overhead is 6% on 1600 cores of an SGI Altix 4700 and less than 7% on a BlueGene/P system at 64Ki cores.