Hewlett Packard Enterprise has built two supercomputers for the United States Air Force's weather research.
Installed at the US Department of Energy’s Oak Ridge National Laboratory, the two HPE Cray EX systems are named “Fawbush” and “Miller” after meteorologists Major Ernest Fawbush and Captain Robert Miller, who made the first tornado forecast at the Tinker Air Force Base in Oklahoma in 1948.
Each supercomputer has a peak performance of 3.6 petaflops, for a total of 7.2 petaflops. Together, that makes them six times more powerful than the previous Thor supercomputer.
Because even tanks don't like rain
The supercomputer contract was awarded to Cray in 2019, ahead of its acquisition by HPE. Both systems feature 2nd Gen AMD Epyc CPUs
“We are thrilled to have built the US Air Force a new supercomputer that is one of the first operational systems powered by the latest HPE Cray EX supercomputer and managed by Oak Ridge National Laboratory (ORNL)," Bill Mannel, VP and GM of HPC at HPE, sais.
"The end-to-end HPC technologies made possible by the HPE Cray EX supercomputer will enable greater speed and dedicated performance to advance simulations in weather forecasting that were never made possible before."
In collaboration with ORNL’s Computational Earth Sciences Division, the Air Force plans to use the new supercomputers to overhaul its weather forecasting capabilities.
In particular, researchers aim to focus on forecasting stream flow, flooding, or inundation to predict how much of a given land will be submerged in water and the level of its depth. Tanks and other heavy military equipment aren't able to travel across soaked ground as fast as dry ground, so predicting the weather can allow military commanders to forecast troop movement.
To do this, the group plans to create a global hydrology model that involves simulating hundreds of watershed and drainage basins to eventually increase accuracy in predicting future events.
For helping flight movement and satellite imagery assessments, researchers also hope to improve remote sensing of a cloud-covered area to address how to navigate impacted missions through forecasting the formation, growth and precipitation of atmospheric clouds. The group expects to do this by using comprehensive cloud physics that are not made possible with existing statistical regression models.
Overall, the Air Force expects to increase the accuracy of its global weather simulations from 17 kilometers between model grid points to 10 kilometers. It also does higher resolution simulations of targeted regions.
