[adrin]

Hello all,

I would appreciate any information you can provide, based on hands-on experience, with examples, on typical CPU-cost differences between the standard Smagorinsky and Dynamic Kinetic Energy SGS models.

(1) Does the K.E. model typically require higher or lower resolution than Smagorinsky? If yes, by how much (just rough estimate)

(2) For a fixed number of meshes, what's the CPU for K.E. vs Smagorinsky (either in serial or parallel mode)

Thanks very much in advance

adrin