A Department of Energy supercomputer that deploys over 27K Nvidia graphics processing units has helped a Georgia Tech University-led team conduct early research into propulsion system development.
The Summit supercomputer, located at DOE’s Oak Ridge National Laboratory in Tennessee, uses Nvidia’s V100 Tensor Core GPUs to run simulations that help researchers study combustion-related phenomena and potentially prevent instabilities such as turbulence, Nvidia said Thursday.
Joe Oefelein, an aerospace engineering professor at Georgia Tech, said the team aims to mitigate such phenomena by maximizing vehicle performance while reducing emissions. The team used Summit to simulate flow movement and create a dataset on turbulence impact and related combustion procedures as part of their first simulation activities.
Oefelein noted that Summit, which has the capacity to run code 25 times faster that a prior system, will now support the team’s efforts to develop predictive capabilities for propulsion systems.
“As we move toward more predictive models, you’re going to see innovative design. We’re going to get improved fuel economy, reduced emissions and high performance,†he said.