A GPU Revolution in Discrete Element Method Applications

Learn about the impact of GPU acceleration on DEM simulations from real-world users at the ATCx Discrete Element Method event.

Learn about the impact of GPU acceleration on DEM simulations from real-world users at the ATCx Discrete Element Method event. Both CNH and Astec Industries will explain how they use Altair EDEM DEM simulation and NVIDIA GPUs to optimize their heavy equipment designs.

CNH leverages DEM simulation to optimize the designs of its agricultural equipment. Image courtesy of CNH.


GPU acceleration has been rapidly adopted by simulation software vendors across multiple physics. Discrete Element Method (DEM) solutions are also beginning to take advantage of GPU capabilities, including Altair incorporating GPU acceleration into its EDEM solution. This has not only reduced computation time, but also increased the ability to conduct larger and more complex simulations with great fidelity.

At the upcoming ATCx Discrete Element Method virtual event Nov. 20-21, GPU maker NVIDIA and Altair will take part in a panel discussion, “Revolutionizing DEM Simulations: The Impact of Next-Gen GPUs and Computational Technologies,” along with end users Astec Digital and CNH.

DEM simulation is a technique for modeling the behavior of granular materials and other particulate systems, and is used in industries ranging from pharmaceutical manufacturing to mining and construction. Computational efficiency has traditionally been a bottleneck in the DEM space. GPU acceleration improves efficiency and memory usage.

CNH is one of the world’s leading agricultural equipment manufacturers, and its products include tractors, combine and specialty harvesters, seeding tools, sprayers, tillage tools and more. According to Ryan Roberge, Soil & Crop Modeling Engineer at CNH, the company has been using the EDEM tool since 2006. “Within the modeling team, we use EDEM as one of our primary tools to assess the flow of crops and soil through agricultural equipment,” he says. “For example, the flow of grain through a combine harvester or soil flow around a seeding knife, or the conveyance of seeds and fertilizer going from a tank to the ground.”

Using DEM simulation, engineers can see material movements that would not be possible with physical testing, and allows designers to make changes that reduce equipment power requirements or reduce damage to seeds and fertilizer products during conveyance. 

Using simulation, the company also calibrates analysis based on different plant types in different geographies. “A wheat crop in the southern U.S. is not the same as one in Europe or Australia,” Roberge says. “With simulation, we can see various materials and how they react to similar system designs. We can also run virtual tests year-round, versus the narrow window of time we have for field testing during different growing seasons.”

Heavy equipment manufacturer Astec Industries also uses Altair EDEM for simulating particle behavior in its asphalt road building and concrete production equipment. Andrew Hobbs, Director of Advanced Technologies at Astec Digital, says simulation allows engineers to see what is happening inside its equipment in order to optimize performance. Hobbs says the software allows the company to simulate particles using both spheres (which are more computationally efficient) and polyhedral shapes, depending on the question they are trying to answer.

An Altair EDEM simulation rendered in NVIDIA Omniverse. Image courtesy of Astec Industries.

At Astec, Hobbs says using NVIDIA RTX A6000 GPUs, Astec testing showed a 90x improvement in simulating realistic particle shapes. “GPU and DEM are made for each other because of the nature of contact detection algorithms. It has been a game changer in terms of the number of particles we can simulate and the computational speed,” Hobbs says.

At CNH, moving from CPU to GPU processing has allowed up to an order of magnitude increase in the number of particles that can be simulated for the same input. The company can run simulations faster, and also run much more complex simulations. 

CNH currently runs the software on workstations with dual CPUs and NVIDIA RTX 6000 Ada GPUs, and plans to transition to the NVIDIA A800 GPU during their next hardware refresh.

“The professional series of GPUs from NVIDIA comes in a two-slot form factor, which is important for high-end workstations, offers great cooling, stability and a high amount of VRAM memory,” Roberge says.

You can learn more about DEM and GPU acceleration at the upcoming panel discussion during the Altair ATCx Discrete Element Method virtual event. You can see details and register here.

More Altair Coverage

Altair Company Profile

More NVIDIA Coverage

Share This Article

Subscribe to our FREE magazine, FREE email newsletters or both!

Join over 90,000 engineering professionals who get fresh engineering news as soon as it is published.


About the Author

Brian Albright's avatar
Brian Albright

Brian Albright is the editorial director of Digital Engineering. Contact him at de-editors@digitaleng.news.

Follow DE
#29588