When it comes to Formula 1, as the old saying goes, speed is everything.
In what is proving to be an eye-opening season of transition and excitement, getting the upper hand on your rivals has never been more important as a new set of rules and regulations has shaken up the grid.
Having won the last eight constructors’ world championships, Mercedes F1, home to seven-time drivers’ champion Lewis Hamilton, has struggled so far this season, trailing far behind Ferrari and Red Bull.
However, the team knows it can still turn to one of the leading IT firms to help raise its game, as it revealed more about its extended partnership with AMD.
The semiconductor company has provided its second-generation EPYC processors to help power the team’s aerodynamic testing, a vital part of developing a potentially race-winning car.
“We are proud to partner with the reigning constructors champions, the Mercedes-AMG Petronas Formula One Team, which operates at the forefront of racing and technology,” said Dan McNamara, senior vice president and general manager of the Petronas Business Unit. AMD servers.
“For F1 teams, having the most effective computational analysis of aerodynamics can mean the difference between winning or losing a race. With AMD EPYC processors, the Mercedes-AMG F1 Team can iterate on vehicle design faster and more efficiently than its previous system.”
In a blog post expanding on the news, AMD pointed out how Mercedes F1 has been able to achieve a 20% performance improvement for the computational fluid dynamics (CFD) workloads that were used to model and test the aerodynamic flow of its F1 car by using their EPYC processors. .
With gigabytes of data resulting from wind tunnel testing and modeling racing, aerodynamics have never been more valuable to a Formula 1 team, especially now that the sport’s governing body, the FIA, has introduced strict rules that govern how many tests can be done in one attempt. to level the playing field between rich and poor teams.
This includes a budget cap of $140 million in computational resource expenditures, which includes money spent on servers, and this amount will drop to just $135 million in 2023. About 1,800 new geometry simulations are allowed in a trial period. eight weeks, so squeezing every piece of information is vital.
Having signed a three-year deal to use second-generation EPYC hardware, Mercedes is now hoping for an uptick in fortunes on the track as Hamilton and teammate (and fellow Brit) George Russell look to improve results.
“AMD EPYC processors give us a platform that delivers streamlined day-in, day-out performance at the highest level possible, while meeting our goal of faster turnaround time for design iteration,” said Simon Williams, director aerodynamic development software for Mercedes-AMG Petronas F1.
“Unbelievably, we saw a 20 percent performance improvement over our previous system, cutting our CFD workload time in half. This is a big step up from the previous one or two percent gains seen with previous systems.”