Formula 1* racing is about speed, precision, and the raw nerve of a team that includes drivers, crew, engineers, and—increasingly—IT guys. Technology and data analytics are now, more than ever, a significant part of the effort it takes to help teams like Caterham F1 gain precious tenths of a second to be more competitive on the track.
Back in the day, I earned an aeronautical engineering degree and studied both aerodynamics and fluid mechanics. Those same disciplines are now using supercomputers and analytics to drive decisions that affect F1 race performance through virtual environments in which the cars and their components are designed, tested, and developed. Dell and Intel have created a fascinating video showing how the Caterham F1 Team uses high-performance computing to make their cars faster and safer.
Source: Caterham F1* Team video
The war of physics: Formula 1 racing
Formula 1 racing has been called a “war of physics” because improving the engineering for a car based on aerodynamic principles is an ongoing process. The faster this can be done, the greater the competitive advantage for the team.
Wind tunnels are an essential way to test car performance by using computational fluid dynamics (CFD) to understand the aerodynamics of the vehicle. As of 2014, F1 regulations restrict physical wind tunnel testing, making a virtual model mission critical. With high-performance computing (HPC), engineers can measure the air flow around the entire car in a virtual wind tunnel and explore how various parameters affect car performance. The team can run hundreds of designs through the virtual wind tunnel to determine which ones work best. These can be manufactured and then tested in a physical wind tunnel.
Fast, precise engineering
The Caterham F1 Team relies on a supercomputer that uses 186 Dell* PowerEdge* servers powered by 1,488 Intel® Xeon® cores to process and analyze the large volume of virtual wind tunnel data. More data is collected from physical wind tunnel runs, the track, and a driver-in-loop simulator, all analyzed to maximize car and driver performance for each race depending on the specific track and conditions.
The supercomputer delivers the results fast with the ability to run 17 trillion operations per second. For Caterham, that translates to results in only 12 hours rather than the five months it would take an average PC.
I find it amazing that computing has come so far as to be able to take all of the variables that are present in this type of scenario and spit back a very accurate read of the aerodynamic properties of a vehicle very quickly. It’s a great way to show that technology can be leveraged to virtually test real-life situations as accurately as, less expensively than, and more rapidly than real physical tests.
Take two minutes and watch how Caterham’s supercomputer helps the team compete, and let me know what you think.
#ITCenter #HPC #CaterhamF1