Meeting 3/24

I'm now helping Eddie run his high res Mach Stems on BlueStreak. They look like they will take ~30 hours ea. on 128 nodes + queue time.

I've written in my approximate Riemann Solvers and reconstructed my code so that it's easier to swap in different solvers in the update stage. Currently, it has the solvers which approximate the star states:

  • Exact
  • Primitive Variable
  • Two Shock
  • Two Rarefaction

and then options that cycle through those routines depending on the state U:

  • Exact Solver
  • Adaptive Iterative
  • Adaptive Non-Iterative.

Unfortunately I'm running into some bugs where my code will crash about halfway through the run, which I think may be related the routine to find Smax (the largest wavespeed) in order to calculate the time step. I'm going to get that ironed out this week and hopefully implement the HLLC solvers in the Flux calculation portion of the code as well.

Attachments (4)

Download all attachments as: .zip

Comments

No comments.