New YTEP needs review: major changes for SPH data
I've just opened a pull request for a new YTEP:
This describes a big piece of work that several community members have been collaborating on for the past year or two. I've personally been involved over the past six months in an effort to get this into a state where it can be included in yt.
The goal is to improve the scaling, performance, and ease of use of yt when working with particle data (including and especially SPH particle data). The main user-facing change is the removal of the global octree. If this YTEP is accepted, yt will return data SPH or N-body particle data in a more "particle-centric" fashion, without interpolating data onto an octree, before presenting it to the user. Note that this only affects users of N-body or SPH particle codes, yt will behave the same when working with particle data from grid or octree AMR codes.
If you are a user or developer of yt who works heavily with particle data, I would very much appreciate your eyes on this document as well as any comments, criticisms, or suggestions you might have.
There has already been significant code development on this front, but we're still early enough that we can make major changes in response to feedback on the YTEP and code.
Finally, if any of you are interested in devoting time to this effort, please let me know! There are still lots of things that are broken and need developer time to fix and I am happy to provide suggestions to start contributing.
Thanks for your time!