United States Government as represented by the Administrator of the National Aeronautics and Space Administration; Lockheed Martin Corporation; The Boeing Company; Intelligent Light; Pointwise, Inc., and The Board of Trustees of the University of Illinois
Climate Impact & Sustainability Data (2020)
Reporting Period: 2020
Environmental Metrics
Climate Goals & Targets
Environmental Challenges
- Most domains have not reached the predicted level of capability.
- Effort required to accomplish technology demonstrations is far beyond what could be used routinely.
- Challenges in transitioning to a new computing paradigm in HPC, including minimizing data motion across complex memory hierarchies and increasing node-level parallelism.
- Improving predictions of separated flow remains a challenge for RANS and hybrid RANS/LES methods.
- Integrated transition prediction has not been met for most critical applications.
- Uncertainty quantification (UQ) in CFD has not proceeded as rapidly as envisioned.
Mitigation Strategies
- Steady, incremental progress in unsteady methods for high Reynolds number flight over complex configurations.
- Significant contributions from time-accurate simulations in the upcoming 4th AIAA CFD High Lift Prediction Workshop.
- Successful porting of NASA FUN3D solver to NVIDIA GPU architectures and effective scalability to 6144 GPUs.
- Development of numerous carefully defined test cases for testing and validation of RANS models and scale-resolving simulations.
- Active research in applying machine learning to improve RANS modeling predictions.
- Advancements in both development of appropriate methods for performing UQ analysis and in applying these methods to CFD analysis.
- Increasing focus on multifidelity techniques using CFD.