Wanted: CFD Mesh and Solution Data For Analysis of Mesh Quality Effects on Solution Accuracy

This site is intended to serve as a repository for data that can be used to study the interaction between mesh quality and solution accuracy. I've set this site up both to collect data that will be useful to me in my own research and to distribute that data to other researchers with similar interests.

If you're here, you probably don't need to be convinced that this is a problem, but in case you do, here's a thumbnail overview, including some thoughts on how such data could be used.

Obviously, for this repository to work, we need data. Here are some guidelines for those interested in contributing data:

To keep the scope of analysis to a reasonable level, solution data should be for steady state and based on either the Euler or Navier-Stokes equations. Both laminar and turbulent viscous flow solutions are welcome; thin-layer approximations are fine, too.
Solution data should be created by CFD solvers that are well-verified and well-validated. If your code isn't producing physically and numerically sensible solutions, you've got worse problems than mesh quality.
The data should not be proprietary, classified, or otherwise restricted in how it can be distributed. By default, solution data will be posted for outgoing FTP access, with one exception, listed below.
Bad solution data is likely to be at least as useful as good solution data! So don't be shy about sending in solutions where you thought everything should go right and it didn't. If the solution is bad enough that you don't want it distributed with your name on it, let me know. I can either distribute it without attribution or not distribute it publicly at all.

This is a repository for sharing research data, not a consulting service. While I'm happy to try to answer questions of the form, "Why can't we get a good answer for this problem? My client / boss needs to know immediately!" that's obviously a consulting question, not a research question. So please contact me directly with questions like this rather than just uploading your data; you're much more likely to get a quick response that way.

How To Upload Data

Each submission should be a self-contained compressed archive (.tgz, .tar.bz2, .zip, etc), including a copy of the appropriate data from, solution data, mesh data, code fragments that will read the solution and mesh data (unless the file format is one commonly read by commercial post-processing software), and any other information about the problem that the data form requests or that you think might be useful. In the very near future, we'll have a set of CGNS-based I/O routines available for use, providing a common file format that won't require anyone to write I/O code to read anyone else's data.

Data for cases from the Fourth AIAA Drag Prediction Workshop should use this data form and be uploaded to ftp://mesh-quality.mech.ubc.ca/pub/incoming/DPW4.
Data for other aerodynamics problems should use this data form and be uploaded to ftp://mesh-quality.mech.ubc.ca/pub/incoming/aero.
Data for other problems should use this data form and be uploaded to ftp://mesh-quality.mech.ubc.ca/pub/incoming/other.

How To Download

All publicly available data is served from ftp://mesh-quality.mech.ubc.ca/pub/outgoing.

Questions / comments? Contact Carl Ollivier-Gooch (cfog@mech.ubc.ca).