Deep and Shallow Convections in Atmosphere Models on Intel® Xeon Phi™ Coprocessor Systems

Deep and shallow convection calculations occupy significant times in atmosphere models. These calculations also present significant load imbalances due to varying cloud covers over different regions of the grid. In this work, we accelerate these calculations on Intel® Xeon Phi™ Coprocessor Systems. By employing dynamic scheduling in OpenMP, we demonstrate large reductions in load imbalance and about 10% increase in speedups. By careful categorization of data as private, firstprivate and shared, we minimize data copying overheads for the coprocessors. We identify regions of false sharing among threads and eliminate them by loop rearrangements. We also employ proportional partitioning of independent column computations across both the CPU and coprocessor cores based on the performance ratio of the computations on the heterogeneous resources. These techniques along with various vectorization strategies resulted in about 30% improvement in convection calculations.

• Publication year

  2017

• Venue

  2017 IEEE 19th International Conference on High Performance Computing and Communications; IEEE 15th International Conference on Smart City; IEEE 3rd International Conference on Data Science and Systems (HPCC/SmartCity/DSS)

• Publication date

  2017-11-01

• Fields of study

  Physics, Computer Science, Environmental Science

• Identifiers
  DOI 10.1109/HPCC-SmartCity-DSS.2017.42arXiv 1711.00289
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Performance tuning Weather Research and Forecasting (WRF) Goddard longwave radiative transfer scheme on Intel Xeon Phi

  2015cited by this paper
• MICA: A fast short-read aligner that takes full advantage of Many Integrated Core Architecture (MIC)

  2015cited by this paper
• Revisiting Intel Xeon Phi optimization of Thompson cloud microphysics scheme in Weather Research and Forecasting (WRF) model

  2015cited by this paper
• Weather Forecasting Using GPU-Based Large-Eddy Simulations

  2015cited by this paper
• Optimizing the updated Goddard shortwave radiation Weather Research and Forecasting (WRF) scheme for Intel Many Integrated Core (MIC) architecture

  2015influential reference
• SWAPHI-LS: Smith-Waterman Algorithm on Xeon Phi coprocessors for Long DNA Sequences

  2014cited by this paper
• Towards a performance portable, architecture agnostic implementation strategy for weather and climate models

  2014cited by this paper
• Optimizing Weather and Research Forecast (WRF) Thompson cloud microphysics on Intel Many Integrated Core (MIC)

  2014cited by this paper
• Lattice QCD with Domain Decomposition on Intel® Xeon Phi Co-Processors

  2014cited by this paper
• Performance Metrics and Application Experiences on a Cray CS 300-AC TM Cluster Supercomputer Equipped with Intel R © Xeon Phi TM Coprocessors

  2013cited by this paper
• Progress towards accelerating HOMME on hybrid multi-core systems

  2013cited by this paper
• Porting Applications with OpenMP Using Similarity Analysis

  2013cited by this paper
• The Mean Climate of the Community Atmosphere Model (CAM4) in Forced SST and Fully Coupled Experiments

  2013cited by this paper
• GPU Acceleration of a Cloud Resolving Model using CUDA

  2012cited by this paper
• An 80-Fold Speedup, 15.0 TFlops Full GPU Acceleration of Non-Hydrostatic Weather Model ASUCA Production Code

  2010cited by this paper
• Running the NIM Next-Generation Weather Model on GPUs

  2010cited by this paper
• HPCTOOLKIT: tools for performance analysis of optimized parallel programs

  2010cited by this paper
• The University of Washington Shallow Convection and Moist Turbulence Schemes and Their Impact on Climate Simulations with the Community Atmosphere Model

  2009cited by this paper
• The Formulation and Atmospheric Simulation of the Community Atmosphere Model Version 3 (CAM3)

  2006cited by this paper
• The Accumulation of Rounding Errors and Port Validation for Global Atmospheric Models

  1997cited by this paper
• Sensitivity of Climate Simulations to the Parameterization of Cumulus Convection in the Canadian Climate Centre General Circulation Model

  1995cited by this paper
• Parallel Processing Letters c ○ World Scientific Publishing Company GPU ACCELERATION OF NUMERICAL WEATHER PREDICTION

  year unknowncited by this paper

• No citing papers are available for this paper.