{"id":4929,"date":"2021-10-13T12:40:17","date_gmt":"2021-10-13T01:40:17","guid":{"rendered":"https:\/\/www.computationalfluiddynamics.com.au\/?p=3712"},"modified":"2025-05-14T15:48:37","modified_gmt":"2025-05-14T04:48:37","slug":"more-gpus-faster-processing-with-rocky-dem-4-5","status":"publish","type":"post","link":"https:\/\/www.leapaust.com.au\/blog\/dem\/more-gpus-faster-processing-with-rocky-dem-4-5\/","title":{"rendered":"Finding the sweet spot with GPU vs CPU hardware for faster processing in Rocky DEM"},"content":{"rendered":"
<\/div>\t\t
\n\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t

At LEAP, we work with many customers who are at the forefront of computational particle modelling \/ discrete element modelling, with engineers and analysts from industries as diverse as minerals\/mining through to agricultural and pharmaceuticals. We\u2019ve certainly seen a request for not only faster solution times but higher fidelity simulations through increased number of particles that customers want to include in their DEM simulations (driven by a desire to more accurately represent particle size distributions, PSDs), as well as increased complexity of the particle shape (using more facets to accurately reflect the true shapes of particles such as rocks and tablets).<\/em><\/p>

Fortunately, we are currently living through a period where computational hardware is evolving at a rapid pace \u2013 over the past 5 years or so, GPU computational capability has increased by several orders of magnitude. We\u2019ve also been fortunate enough to be working with ESSS who develop Rocky DEM<\/a> which was the first commercial DEM code to offer a dedicated GPU-based solver and later extending this to multiple GPUs. Developers at ESSS<\/a> had also been observing the same level of interest from Rocky DEM customers worldwide and recently reviewed this topic in detail where they completed a comprehensive set of benchmarks that involved comparing the solve time on an 8-core CPU against single and multi-GPUs \u2013 you can see the results summarised here:<\/em><\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t

\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\"\"\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t
\n\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t

In the past, DEM simulations were restricted to relatively small problems that used, for example, only thousands of larger particles that were mostly spherical in shape.<\/p>

Continual improvements in both DEM codes and computational power have enabled closer-to-reality particle simulations. Users today can expect to simulate problems using the real particle shape and the actual particle size distribution (PSD), creating DEM simulations with many millions of particles.<\/p>

However, these enhancements in simulation accuracy have come at the cost of increased computational loads in both processing time and memory requirements. Within Rocky DEM, these loads can be offset considerably by using GPU processing abilities<\/strong>, which provides users with the capacity to obtain results in a more practical time frame.<\/p>

The benefits of GPU<\/strong>
The addition of GPU processing has helped make DEM a practical tool for engineering design. For example, the speed-up experienced by processing a simulation with even an inexpensive gaming GPU is remarkable when compared to a standard 8-core CPU machine working alone.<\/p>

Since release 4 of Rocky, users have been able to make use of multi-GPU technology capabilities, which facilitates large-scale and\/or complicated solutions that were previously impossible to tackle due to memory limitations. By combining the memory of multiple GPU cards at once, users have been able to overcome these limitations and achieve a substantial performance increase by aggregating their computing power.<\/p>

From an investment perspective, there are many benefits to multi-GPU processing. The hardware cost of running cases with several millions of particles using multiple GPUs is much smaller than buying an equivalent CPU-based machine. The energy consumption is also less with GPUs, and GPU-based machines are also easier to upgrade by adding more cards or buying newer ones.<\/p>

Moreover, in a world where we push multi-physics simulations ever farther, Rocky GPU and multi-GPU processing enables you to free-up all your CPUs for coupled simulations, avoiding hardware competition.<\/p>

Performance benchmark<\/strong>
To better illustrate the gains in processing speed that are possible for common applications, a performance benchmark of a rotating drum (Figure 1) was developed. Multiple runs using different criteria were evaluated as explained below.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t

\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\"\"\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t

Figure 1 \u2013 Rotating drum benchmark case for spheres (left) and polyhedrons (right).<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t

\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t

Criteria 1: Particle shape<\/strong>
Two different particle shapes were evaluated at the same equivalent size (Figure 2):<\/p>