Data structure for galaxy cluster data

[bnord]

How should we design our data structure for haloes and clusters? This applies to SZ, Optical, and X-ray.

Some options that I can think of:

1. Halo class that has properties (attributes) of both the DM halo and the cluster.
2. A set of lists: e.g., halo ID, halo mass, SZ signal

[kbanker]

Quick clarification, is this about use within the sim/in code, or for the outputs in h5?

[bnord]

both

[bnord]

I could imagine storing a list of halo objects in the h5. I think h5 permits that, but I'm not sure.

[simone-shev]

h5 doesn't actually permit that but there are workarounds