We are delighted to announce our new release – PreonLab 6.1. It is loaded with new features and improvements, enabling new engineering possibilities and applications. Empowered by the new GPU capabilities, PreonLab 6.1 takes another leap forward in efficient computing, delivering results even quicker. Read on to find out more about:
This is just a selection of new features and improvements. Check out the changelog to learn about all the changes.
Make sure to follow us on LinkedIn so that you don’t miss new videos, case studies and updates!
While GPU support may be the most important single lever for increased performance, it is not the only one. PreonLab 6.1 comes with continuous particle size, and dynamic and adaptive sampling, extracting the most out of your GPU hardware.
These features cleverly reduce the number of fluid and solid particles in the scenes which translates to reduced memory requirements. With these features, the benefits of simulating on GPU with PreonLab are further improved – enabling the simulation of new applications or simulating existing ones more efficiently and accurately. With the availability of these features, all important efficiency-improving features available on the CPU platform are now also ready to be used on the GPU. Go ahead and use them, they all make a world of difference.
What about scaling to multiple GPUs?
Optimizing our code, fine-tuning data structures, and porting key features first were integral steps in our plan to implement multi-GPU support. Today, we’re pleased to announce that PreonLab 6.1 also features single-node multi-GPU support.
The addition of multi-GPU support expands available memory by leveraging multiple GPUs simultaneously. The synergy between memory saving features (CPS, dynamic sampling, adaptive sampling) and the multi-GPU support make it possible to execute simulations not possible before due to memory constraints. Do you want to simulate a large domain? Easy. Do you need a super fine resolution? No problem. Simulating with PreonLab on GPU was never easier than today.
Solving the memory bottleneck paved the way for porting the application-specific features. With airflow support available on GPU, one-way coupled simulations including airflow are now possible to be simulated. One of the more demanding applications in this field is soiling. By leveraging the power of GPU, these use cases can now be more easily simulated with reasonable turnaround times.
Vehicle water wading has been historically one of the most computationally expensive use cases. With resolved memory constraints and a car suspension model on GPU, this application is one of the clear winners that will benefit the most from new GPU capabilities.
Our elastoplastic snow model is next on the list of application-specific features now available on GPU. Calculating elastic and plastic deformations can be computationally expensive especially when there is a need for fine resolution mimicking snowflakes. This makes these snow-related applications a perfect candidate for GPU-powered PreonLab.
The release of PreonLab 6.1 brings significant upgrades to its GPU implementation. But this is not all since it includes also numerous platform-independent upgrades. This latest version introduces features such as pathlines with arrow indicators, batch video processing, a redesigned coordinate system, and additional camera adjustment options.
Check out the changelog for a full list of changes. To learn more about the new features, have a look at the updated manual. We hope you will enjoy working with PreonLab 6.1 and are looking forward to your feedback.