#UFOCFD cartesian, 20 million cells, 50k iterations, 32 minutes runtime on dicehub GPU #FordvsFerrari model geometry from thingiverse.com/thing:4738286

#UFOCFD (octree) #drivaer geometry at 100m/sec, 2million cells, 40k iterations, 5 mins runtime dicehub GPU

dicehub (dicehub): A cloud-based platform that simplifies engineering simulations startupbase.io/startups/diceh…

#UFOLBM Ferrari Dino case at Re=3000 (2D slice) mesh 100k cells runtime 40 seconds - Ran on single GPU via the dicehub platform. Now to extend the method for higher Reynolds numbers using the regularized RLBM approach.

#UFOLBM motorbike - 32M cells (Re=1million), ran on dicehub single GPU in 2 hours

🎄✨ From prompt to 3D to CFD to fun art! 🦌🚀 Happy Christmas!

Analyze Car Aerodynamics in Minutes with dicehub! 🚗💨 Upload your car design, let dicehub set up the wind tunnel, choose velocity, and start your simulation. Get CFD results in no time! #CFD #Aerodynamics #dicehub #SimulationMadeSimple #EngineeringInnovation

#UFOLBM Porsche demo - 100M cells ran for 40k iterations in 3.25 hours on dicehub single T4 GPU. Here testing a new wall 'reflection' boundary condition approach for D3Q15 LBM. All post-processing done in Kitware #Paraview.

#UFOLBM Ferrari Dino demo 100M cells ran on dicehub single T4 GPU in 3.25 hours. Here testing again the new wall reflection BC which allows us to specify higher velocity at the wall (compared with the bounce-back method). I put some details here - docs.google.com/presentation/d…

Coming Soon: The allpccloud FEM Solver on dicehub! 🚀 Get ready to elevate your engineering workflow with the upcoming allpccloud FEM Solver, soon to be seamlessly integrated into the dicehub platform. Boost your productivity and simplify your simulations like never before. Why

#interFoam annular extrusion onto a moving cylinder - this was run on dicehub using the #caserun facility to upload your #openfoam project directory and run it on multiple processors in the cloud - then download the results when its finished.

#interFoam floating Buddha ran on dicehub

#interFoam floating Buddha ran on dicehub

#UFOLBM ran on dicehub single T4 GPU. Cartesian mesh 72 million cells. Re=0.5 million flow over sphere. 100k iterations took 5.6 hours runtime, saving every 1000 iterations. This baseline testcase is using the 'bounce-back' wall BC, hence large separation region.

#UFOLBM 3D flow over sphere with different wall treatments. Mesh 72 million cells run on dicehub single T4 GPU for 100k iterations and taking 5.5 hours each. The testcase is Re=0.5million for 10m/sec flow. Top left shows bounce-back, the others with different wall fn settings.

#UFOLBM with "bounce-back" wall treatment and setting Re=500,000. This case is 96 million cells and took 5.85 hours to run on dicehub single T4 GPU grabcad.com/library/sports…

#UFOLBM Ford GT40 at 30m/sec - 96 million cells 200k iterations. Ran on single T4 GPU via dicehub, runtime 15 hours. VTK files saved every 1000 iterations. Here showing eddy viscosity contours. (Re = 1.8million)

#UFOLBM Ford GT40 at 30m/sec - 96 million cells 200k iterations. Ran on single T4 GPU via dicehub, runtime 15 hours. VTK files saved every 1000 iterations. Here showing velocity magnitude contours (Re = 1.8million).

If a teenager can do this, just imagine what an engineer with a bit of experience could pull off. Kudos Ted. dicehub - Create high-fidelity fluid dynamics simulations effortlessly, in just a few clicks. #openfoam #aerodynamics #cfd bit.ly/4lx0fjQ

#rhoSimpleFoam F1 car, 100m/sec, ran in parallel on dicehub (18 processors), 1000 iterations, 1.8M cells