Parallelizing TriScatteredInterp() on ANSYS Fluent output data
2 views (last 30 days)
Show older comments
Howdy. I am trying to generate contour/vector plots for about 15 thousand time steps of a transient computational fluid dynamics solution. I've successfully used TriScatteredInterp to interpolate scattered nodal data onto the output of meshgrid/ndgrid, but the operation is time consuming (I start with about 52 thousand nodes, and that's just on the plane I am generating plots for -- if I was doing this in 3D there would be millions). By "time consuming" I mean it is taking on the order of several seconds (maybe ten?) for a single time step. I have twelve cores to play with, so I can cut it down by an order of magnitude if I use parfor (does anyone have any experience with this?). But I was really hoping that I could use the GPU (Quadro FX 3800, 192 processors) to do this. However, it appears that this is not supported ( http://www.mathworks.com/help/toolbox/distcomp/bsic4n4-1.html ). Does anyone have any suggestions as to how I can significantly cut down the time required for my post-processing? Thanks!
1 Comment
Answers (0)
See Also
Categories
Find more on Performance Profiling in Help Center and File Exchange
Products
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!