Never miss a story from Computational Fluid Dynamics. Subscribe for in-depth analysis and articles.
As a car enthusiast, you are likely aware of the constant evolution of the automotive industry. The advent of hypercars has driven significant changes in how cars are designed, built, and driven. But have you ever considered what goes on behind the scenes of the racing world? Revving engines, screeching tires, and the thrill of speed—that’s what keeps us glued to our seats. But behind the roar of the engines lies a world of innovation and precision engineering. One of the latest and greatest technologies in an automotive designer's arsenal is GPU acceleration, to design cars that run faster, lighter, and cooler than ever. In this article, we'll look closer at the significance of GPU acceleration in automotive design and technology. We will also discuss the latest developments in the racing world, including McLaren's progress towards creating a fully sustainable ‘circular’ F1 car. So, fasten your seatbelts and get ready for an exciting ride!
The latest performance of the McLaren team has left the racing enthusiasts awestruck! The team has put up an impressive show on the high-downforce tracks at the Qatar Grand Prix, with the racers Oscar Piastri and Lando Norris finishing the race incredibly close to Max Verstappen from team Red Bull. The McLaren team has shown remarkable progress in recent races and has performed exceptionally well in the Japanese, Singaporean, Hungarian, and British Grand Prix. Lando Norris' consistent second position hold during these races has been a sight to behold! The excitement levels are high for the upcoming United States Grand Prix on October 23. We can't wait to see what the McLaren team has in store for us!
Improving the racing performance of a team can be partly credited to the vehicle’s aerodynamic design. Aerodynamic design discussions are incomplete without mentioning the importance of computational fluid dynamic (CFD) simulations, especially in the current competitive times. Restrictions on the use of GPUs and limited CPU cores for CFD testing in Formula One highlight the need for versatile CFD platforms that can cater to a wide range of customer applications. While some CFD applications might just require RANS simulation, which can be run using a single-core CPU resource, others might require GPU-accelerated LES simulations to capture the flow features precisely.
GPU-accelerated CFD solutions can be a boon for racing teams outside Formula One. GPU-acceleration and GPU-resident solvers allow the exploration of complex design spaces with unprecedented speed and accuracy. It is fundamentally changing the game. The graph below compares the simulation throughput of the Cadence Fidelity CharLES Solver performed on CPU and GPU. For a given usage of compute, measured on AWS pricing of CPU and GPU nodes, the NVIDIA GPUs deliver nearly nine times the throughput of the CPU.
NVIDIA’s CEO, Jensen Huang, in his keynote at Computex 2023, mentions NVIDIA being a major customer of Cadence that uses all our tools that run on CPUs. Most of their algorithms are very CPU-centric. They have been working on accelerating some of these algorithms with Cadence for some time, but with the Grace Hopper Superchip, the performance speed-up has been incredible, and we can't wait to share the results. This breakthrough superchip can potentially revolutionize industries requiring high levels of computation, such as computer-aided engineering (CAE), chip design, and digital biology.
Along these lines, having a bundle that improves CFD simulations by enhancing the numerical scheme and enabling GPU acceleration would be disruptive. Moreover, CFD users are also on the lookout for the capability to allow easy upgrades to the simulations in the same environment faster equipment becomes available.
As we discuss the automotive simulation needs and trends, you might be interested to know about the recent CadenceCONNECT event, "The Race is On!” that took place on October 18 at Brazos Hall in Austin. The event provided a broad perspective on automotive technology, highlighting the latest advancements in automotive chip design, including upcoming trends and innovations such as autonomous driving, electric vehicles, functional safety, and reliability of automotive semiconductors. The event also showcased a new generation of automotive electronics chip design full-flow solutions.
For students, we have an exciting internship lined up with Cadence and the McLaren F1 team. All you need to do is solve for the optimal downforce-to-drag ratio of the front wing of a race car, and your entry could qualify you for a Cadence 2023 internship. The final deadline for submission of the Team CFD Student challenge design is February 16, 2024, at 12:00pm PST.
The world of Formula One is always pushing the boundaries of technology and innovation, and McLaren is again leading the charge in this regard. The team recently announced that it will be testing cutting-edge recycled carbon fiber (rCF) on Lando Norris and Oscar Piastri's cars, i.e., on the cockpit branding panels, at the upcoming United States Grand Prix. This initiative is part of McLaren's larger goal to create a fully circular F1 car by 2030 while also helping to identify and test new materials and technologies on the road to achieving its NetZero targets. Seeing McLaren taking bold steps towards a more sustainable future for racing and beyond is exciting.
As we wrap up this article, it's clear that the automotive industry is hurtling toward a thrilling and innovative future. We're talking electric cars, autonomous driving systems, and mind-blowing simulations that will push the boundaries of what we thought was impossible. The racing world is undergoing a revolution, thanks to the latest developments in technology and design. With the help of GPU acceleration, the industry is making quantum leaps in what it can achieve.
If you would like to try Cadence Fidelity CFD for your automotive applications, request a demo today!