Abstract
Weather models with high spatial and temporal resolutions are required for accurate prediction of meso-micro scale weather phenomena. Using these models for operational purposes requires forecasts with sufficient lead time, which in turn calls for large computational power. There exists a lot of prior studies on the performance of weather models on single domain simulations with a uniform horizontal resolution. However, there has not been much work on high resolution nested domains that are essential for high-fidelity weather forecasts.
In this paper, we focus on improving and analyzing the performance of nested domain simulations using WRF on IBM Blue Gene/P. We demonstrate a significant reduction (up to 29%) in runtime via a combination of compiler optimizations, mapping of process topology to the physical torus topology, overlapping communication with computation, and parallel communications along torus dimensions. We also conduct a detailed performance evaluation using four nested domain configurations to assess the benefits of the different optimizations as well as the scalability of different WRF operations. Our analysis indicates that the choice of nesting configuration is critical for good performance. To aid WRF practitioners in making this choice, we describe a performance modeling approach that can predict the total simulation time in terms of the domain and processor configurations with a very high accuracy (< 8%) using a regression-based model learned from empirical timing data.
Chapter PDF
Similar content being viewed by others
Keywords
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.
References
Bhatele, A., Gupta, G.R., Kale, L.V., Chung, I.H.: Automated Mapping of Regular Communication Graphs on Mesh Interconnects. In: HiPC (2010)
Delgado, J., et al.: Performance Prediction of Weather Forecasting Software on Multicore Systems. In: IPDPS, Workshops and PhD Forum (2010)
Kerbyson, D.J., Barker, K.J., Davis, K.: Analysis of the Weather Research and Forecasting (WRF) Model on Large-Scale Systems. In: PARCO, pp. 89–98 (2007)
Kumar, S., et al.: The Deep Computing Messaging Framework: Generalized Scalable Message passing on the Blue Gene/P Supercomputer. In: ICS 2008 (2008)
Kumar, S., et al.: Architecture of the Component Collective Messaging Interface. IJHPCA 24(1), 16–33 (2010)
Kutner, M.H., Nachtsheim, C.J., Neter, J.: Applied Linear Regression Models, Fourth International edn. McGraw-Hill (September 2004)
Michalakes, J.: RSL: A Parallel Runtime System Library For Regional Atmospheric Models With Nesting. Tech. Rep. ANL/MCS-TM-197, Mathematics and Computer Science Division, Argonne National Laboratory, Argonne (1997)
Michalakes, J., et al.: WRF Nature Run. In: SC, p. 59 (2007)
Porter, A.R., et al.: WRF code Optimisation for Mesoscale Process Studies (WOMPS) dCSE Project Report (June 2010)
Shainer, G., et al.: Weather Research and Forecast (WRF) Model Performance and Profiling Analysis on Advanced Multi-core HPC Clusters. In: 10th LCI ICHPCC (2009)
Skamarock, W.C., et al.: A Description of the Advanced Research WRF version 3. NCAR Technical Note TN-475 (2008)
Wright, N.J., Pfeiffer, W., Snavely, A.: Characterizing Parallel Scaling of Scientific Applications using IPM. In: 10th LCI International Conference on High-Performance Clustered Computing (2009)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2012 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Malakar, P. et al. (2012). Performance Evaluation and Optimization of Nested High Resolution Weather Simulations. In: Kaklamanis, C., Papatheodorou, T., Spirakis, P.G. (eds) Euro-Par 2012 Parallel Processing. Euro-Par 2012. Lecture Notes in Computer Science, vol 7484. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-32820-6_80
Download citation
DOI: https://doi.org/10.1007/978-3-642-32820-6_80
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-32819-0
Online ISBN: 978-3-642-32820-6
eBook Packages: Computer ScienceComputer Science (R0)