Background and Challenge
Richard Childress Racing is one of the most storied auto racing teams in NASCAR history. Since 1969, RCR has won 17 championships and 200 races and became the first team to win in all three of NASCAR’s top touring series. With six Cup Series championships, RCR is an elite organization. Today, RCR has eight full-time race teams, 500-plus employees, and a 40-acre campus with an engineering staff of over 50. They design and build all their race cars end-to-end. They engineer and machine their own chassis and suspension components, design and fabricate their own bodies, and test and build their own engines.
Today, RCR has eight full-time race teams, 500-plus employees, and a 40-acre campus with an engineering staff of over 50. They design and build all their race cars end-to-end. They engineer and machine their own chassis and suspension components, design and fabricate their own bodies, and test and build their own engines.
Simulation is a critical ingredient to RCR’s success, especially for the aerodynamics team, where small details can make the difference between winning and losing a race. They rely on computational fluid dynamics (CFD) simulation for 95% of the car geometry to develop a deeper understanding of the aerodynamic system of the car, evaluate new aerodynamic concepts, recommend design changes to be tested by the wind tunnel team, and analyze phenomena not apparent in the wind tunnel.
CFD is a compute-intensive process, and RCR has limited on-premise high-performance computing (HPC) infrastructure and in-house expertise to support their simulation requirements. Feeling constrained, RCR’s aerodynamics team wanted to augment their compute capacity to build larger CFD models with a resolution high enough to precisely understand the intricate flow details that affect the car’s speed on the track.
The Rescale Solution
Full-body aerodynamic analysis requires that models be quite large at up to 400 million cells, which takes thousands of compute cores to process in a reasonable time. Bringing HPC in-house was infeasible due to the capital investment and HPC expertise required to set up and manage it. Additionally, RCR recognized that computing hardware performance improves significantly year-to-year and did not want to be locked into using an outdated system after a couple of years. RCR considered various HPC options but felt that Rescale’s wide hardware selection and pay-as-you-go pricing model would scale with their needs, giving them the flexibility to start small and ramp up to full production.
“We’re able to run jobs instantly. I don’t have to wait in a queue, and I can run more simulations and larger simulations because I have access to nearly-unlimited hardware. I like the instant gratification that Rescale gives me. It allows me to debug or run design variations easily.” Seth Morris, Aerodynamicist
RCR uses ANSYS Fluent, which is natively-integrated, benchmarked, and optimized for the cloud on Rescale’s ScaleX® platform, to optimize the aerodynamics of the car and other fluid flows. Fluent is a robust, commercially-available CFD solver with the capability to solve problems in many different flow regimes. RCR use Fluent for more than just external aerodynamics— they also use it to simulate downforce variation, fuel flow, and heat exchange.
On the Rescale platform, RCR easily completes their large jobs on 512 cores or more. They can run more complex simulations in greater quantity because of virtually unlimited horizontal scaling of hardware available on Rescale. Additionally, in an industry defined by speed, it’s a boon to be able to run jobs instantly without waiting in a queue. Not only does this accelerate their simulations, but it gives them the agility to debug and run design variations and “what-if” scenarios on the turn of a dime. These competitive advantages are critical to RCR’s on-track success
The ability to easily and efficiently explore the design space using Rescale has been liberating for RCR. To build on that success, RCR is planning to automate more simulations with Rescale in order to increase throughput and streamline their whole design-to-manufacture workflow. At the production level, the addition of unlimited CFD capacity will be game-changing. This will allow RCR to tailor drag and downforce of the body design depending on the rules and regulations and constraints of each race track in order to optimize for lap time. Advanced, high-resolution CFD, requiring thousands of cores, is extremely useful for understanding the complex aerodynamics of the race car. The car operates as a system, and the compute capacity to run advanced CFD will give them the flexibility to rapidly introduce and optimize different body designs as constraints change on a weekly basis.
“Having on-demand access to HPC through Rescale will give us a competitive advantage that will ultimately lead to more on-track wins.” Seth Morris, Aerodynamicist
In summary, running ANSYS Fluent CFD software on the Rescale cloud platform has enabled RCR’s aerodynamics team to:
- Dramatically reduce time to results by running jobs on hundreds or thousands of cores at once without waiting in a queue. Fluent scales linearly, for example, reducing runtime for a 200 million cell model from 16 hours on 512 cores to under 9 hours on 1,024 cores.
- Simulate more advanced physics and flow details. Though they use a very fine mesh in many high gradient regions around the car, their mesh continues to grow to capture ever-smaller flow details.
- Debug, run design iterations and “what-if” scenarios rapidly.
- Tailor drag and downforce depending on the rules and regulations of each race track to win races. Traffic and drafting analyses consume a large number of resources and take a long time to converge.