Download builder

Pick a product family, cosmologies, file contents, and snapshots. The terminal on the right builds the full URL list and a ready-to-run wget script as you go. By default the script recreates the server's directory layout on your machine, so the cosmology stays in the path and files from different cosmologies or snapshots never overwrite one another:

Untick mirror server dirs (at the bottom of the terminal) to drop everything into a single folder instead.

When your selection is ready, click Download .sh to save download_sahyadri.sh, move it to wherever you want the data, and run bash download_sahyadri.sh. Files download one at a time; for a long list, consider running it in the background (e.g. with screen). If it's interrupted, just rerun it — already-downloaded files are skipped and it picks up where it left off.

Directory layout

Everything lives under the base URL http://sahyadri.tifr.res.in/data/sahyadri_sims/. The first split is halos/ (Rockstar catalogs) vs sims/ (particle data). Orange tokens are placeholders.

Available cosmologies sim_cosmo

File contents

Halo catalogs are gzip-compressed FITS binary tables (one row per halo, matched by id/haloID across files). Particle files are algorithmically compressed and require the Sahyadri reader. Expand a file to see its fields.

Halo catalogs FITS.gz · astropy

Particle subsamples HDF5 · compressed

Reader functions

FITS catalogs read with standard tools. Particle files use compressed blocks and need the reader from the sahyadri-sandbox repository.

Halo catalogs — FITS

# pip install astropy
from astropy.table import Table

def load_halo_catalog(filename):
    """Load a Sahyadri halo catalog (basic / extended / vahc / tens).
    Works directly on the .fits.gz file; returns an astropy Table."""
    return Table.read(filename)        # HDU 1, binary table

t = load_halo_catalog("out_100_basic.fits.gz")
print(t.colnames, len(t))
mvir = t["Mvir"]                      # columns by name

Particle subsamples — compressed HDF5

# git clone https://github.com/a-paranjape/sahyadri-sandbox.git
# then make its readers importable (add the repo to PYTHONPATH)
import numpy as np
from readers import SnapshotReader

def load_particles(sim_stem, snap, subsamples=[1], real=1,
                   blocks=("ids","pos","vel","potential")):
    """sim_stem e.g. 'sims/sahyadri/default2048/'. Returns a dict of arrays."""
    sr = SnapshotReader(sim_stem=sim_stem, real=real, snap=snap,
                        read_header=True, use_compressed=True)
    return {b: sr.read_block(b, subsamples=subsamples) for b in blocks}

data = load_particles("sims/sahyadri/default2048/", snap=100, subsamples=[1])
for k, v in data.items():
    print(k, v.shape)

Snapshot ↔ redshift ↔ scale factor

101 snapshots, 0–100. Snapshot 100 is z = 0 (a = 1). Redshift is computed as z = 1/a − 1. You can download the outputs100.txt file which list the scale factor of the 100 snapshot, the first row entry corresponds to snapshot=0