Formula 1 Redesigns Car for Closer Racing and More Exciting Fan Experience by Using AWS HPC Solutions

The name Formula 1 (F1) is synonymous with speed. F1 has the fastest regulated racing cars in the world, which race in the annual international FIA Formula One World Championship. Recently, F1 needed to redesign its cars to reduce the impact of the turbulent wake that limits the speed and acceleration of following cars. “Ultimately, we wanted to reduce the wake so spectators can see more overtakes and closer wheel-to-wheel action,” says Pat Symonds, chief technical officer for Formula 1.

To redesign its cars, F1 needed to move away from its on-premises computational fluid dynamics (CFD) technology environment, which it uses along with wind tunnel testing to design and test the aerodynamic properties of cars. “We wanted more fidelity in terms of simulation detail for designing our cars,” Symonds says. “Also, we needed more scalability and we wanted to reduce our simulation cycle, so we could perform more complex CFD simulations while decreasing overall execution time.”

F1 moved its CFD simulation environment to a high performance computing (HPC) platform on Amazon Web Services (AWS) to solve its challenges. The company uses Amazon Elastic Compute Cloud (Amazon EC2) instances to run powerful, complex simulations that visualize the wake turbulence of cars and the impact on following cars. F1 chose to use a combination of Amazon EC2 C5n instances and the new AWS Graviton2-based C6g instances to gain maximum flexibility in terms of pricing and performance, based on the job. Additionally, F1 used AWS ParallelCluster to automate provisioning of different HPC clusters for job optimization.

By using AWS, F1 has reduced CFD simulation time by 80 percent, from 60 hours to 12 hours. “We can now set a simulation to run in the evening and have the results the next morning,” Symonds says. “With faster results, we can perform more simulations overall and more quickly get to the final design of the car.”

By relying on AWS Graviton2-powered Amazon EC2 instances, F1 has already lowered the cost of running its CFD workloads by 30 percent and anticipates even more savings. “We think our cost savings could grow to 40 percent,” says Symonds.

Ultimately, F1 is no longer limited by its on-premises CFD environment. “We can now run huge models, with more than half a billion cells,” says Symonds. “This is only possible because of AWS.”

By running its CFD platform on AWS, F1 has designed a car that reduces downforce loss in wheel-to-wheel racing from 50 percent to 15 percent, so a car is much less impacted by the wake of the car in front of it. As a result, F1 can better support its strategic priorities of increasing competitiveness and unpredictability on the track, as well as producing a world-class spectacle for fans. “Now, cars can drive closer to each other and overtake more easily,” Symonds says. “The experience for fans will prove to be much more exciting.”