2012 Agenda

9/12/2012(Wednesday)Session ChairSpeakerOrganization
8:30 Workshop Introduction and Opening Remarks Chris Kerr    
8:40 Session 1- Challenges of Using Hybrid Multi-core Systems Rich Loft    
  The GFDL FMS: Can We Get There From Here?   Jeff Durachta NOAA
  Design of Ocean and Ice Models for New Architectures   Phil Jones LANL
  Kepler GPU Architecture and Benefits to Earth System Modeling   Stan Posey NVIDIA
  Execution Models for the Exascale EraM   Nicholas Wright NERSC
10:00 Discussion Period 1      
10:30 Morning Break      
10:50 Session 2 - Using CUDA for Acceleration Bill Putman    
  Porting COSMO to Hybrid Architectures   Peter Messmer NVIDIA
  Porting the Community Atmospheric Model - Spectral Element Code to Utilize GPU Accelerators   Matt Norman ORNL
  Progress in Adapting the GEOS-5 GCM to CUDA FORTRAN: Successes and Challenges   Matt Thompson SSAI/NASA/GMAO
11:50 Discussion Period 2      
12:20 Lunch - NCAR Cafeteria      
1:20 Session 3 - Programming Tools - Compilers and Programming Models Ilene Carpenter    
  KernelGen - A prototype of auto-parallelizing Fortran/C compiler for NVIDIA GPUs   Dmitry Mikushin University of Lugano
  New Developments in the F2C-ACC Compiler   Mark Govett NOAA
  FORTRAN GPU Compilers: Improving, But No Silver Bullet   Tom Henderson NOAA
2:20 Discussion Period 3      
2:50 Afternoon Break      
3:10 Session 4 - Using OpenMP to Accelerate Applications Mark Govett    
  Multi-Core and NEMO   Mike Ashworth STFC Daresbury Lab
  Hybrid Strategies for the NEMO Ocean Model on Many-core Processors   Italo Epicoco and Giovanni Aloisio University of Salento & CMCC
  Experiences Porting NOAA Weather Models FIM and NIM to new Many Integrated Core (MIC) Architecture from Intel   Jim Rosinski NOAA
3:50 Discussion Period 4      
4:20 Workshop First Day Summary Rich Loft    
4:50 Adjourn      

9:00 Session 5 - Debugging on Heterogeneous Multi-core Systems Bill Putnam    
  Multi-Core Debugging Techniques   Ed Hinkel Rogue Wave Software
  Empowering Modeling and Simulation in the Atmospheric Sciences via Developer-Centric Tools for Heterogeneous Computing Architectures   Ian Lumb Allinea
9:40 Discussion Period 5      
10:10 Morning Break      
10:30 Session 6 - Application Load Balancing Ilene Carpenter    
  Large-scale Simulation of the Global Shallow Water Equations on Heterogeneous CPU-GPU Platforms   Lin Gan Tsinghua University
  A Promising Approach to Dynamic Load Balancing of Weather Forecast Models   Jairo Panetta INPE/CPTEC
11:10 Discussion Period 6      
11:40 Lunch - NCAR Cafeteria      
12:40 Session 7 - Using OpenMP and OpenACC to Accelerate Applications Mark Govett    
  Using OpenMP and OpenACC to Accelerate Applications   Pete Johnsen Cray, Inc.
  Using the Cray Programming Environment to Convert an all MPI code to a Hybrid-Multi-core Ready Application   John Levesque Cray, Inc.
  Improving Performance of Weather and Climate Applications on Multi-core Processors   John Michalakes NREL
  A Hybrid GPU Implementation of the Cubed-Sphere Finite-Volume Dynamics in GEOS-5   Bill Putman NASA
2:00 Discussion Period 7      
2:30 Workshop Summary and Future Plans Chris Kerr    
3:00 Adjourn