SDRBench — NYX

Cosmology: adaptive-mesh hydrodynamics + N-body simulation (Nyx), 512 x 512 x 512, single precision, little-endian.

Raw binary arrays, not an Arrow/Parquet dataset. datasets.load_dataset("sdrbench/nyx") returns the table of files (files.jsonl: path, dtype, shape, sha256). Read the arrays with the sdrbench package or hf_hub_download + numpy as shown below.

This repository is an unmodified mirror of the NYX data of SDRBench, the Scientific Data Reduction Benchmark. The originals are hosted by Argonne National Laboratory on Globus; every archive was unpacked and its files uploaded byte-for-byte (sha256 verified) (revision e6ce039140). metadata/ holds the original SDRBench property and template files. Kept in sync automatically by szcompressor/sdrbench.

• Data provider: Zarija Lukic (LBNL)
• Size: 6.4 GB in 2 variant(s)
• Format: raw little-endian binary, C order (slowest dimension first); dtype and shape per field below

Usage

pip install "sdrbench[sz3]"     # sdrbench + pysz (SZ3)

import numpy as np
import sdrbench
from pysz import sz, szConfig, szErrorBoundMode

ds = sdrbench.dataset("nyx")                # default variant "original"
print(ds.fields)
x = ds["baryon_density"]                     # numpy array, dtype <f4, shape (512, 512, 512)

# compress with SZ3 (pysz) at a 1e-3 value-range-relative error bound
conf = szConfig()
conf.errorBoundMode = szErrorBoundMode.REL
conf.relErrorBound = 1e-3
compressed, ratio = sz.compress(np.ascontiguousarray(x), conf)
y, _ = sz.decompress(compressed, x.dtype.type, x.shape)
max_err, psnr, nrmse = sz.verify(np.asarray(x), y)
print(f"ratio {ratio:.1f}x, PSNR {psnr:.1f} dB, max error {max_err:.3g}")

Files are fetched on first use (in parallel with ds.download("data/")) and cached, pinned to the revision of the installed sdrbench release; if Hugging Face is unreachable the package falls back to the original archive on Globus. Existing local SDRBench copies can be used with sdrbench.dataset("nyx", root="/path").

Without the package, with huggingface_hub and numpy only:

from huggingface_hub import hf_hub_download
import numpy as np
p = hf_hub_download("sdrbench/nyx", "512x512x512/baryon_density.f32", repo_type="dataset", revision="e6ce03914098d6dcdf160cb3fe567e7c784c6edf")
x = np.fromfile(p, dtype="<f4").reshape((512, 512, 512))

Variants and fields

original (default)

sdrbench.dataset("nyx", "original") — folder 512x512x512/ — 7 files, 3.2 GB, from SDRBENCH-EXASKY-NYX-512x512x512.tar.gz.

Six fields: baryon and dark-matter density, temperature, velocities.

Field	dtype	Shape (C order)	File
baryon_density	<f4	512 x 512 x 512	512x512x512/baryon_density.f32
dark_matter_density	<f4	512 x 512 x 512	512x512x512/dark_matter_density.f32
temperature	<f4	512 x 512 x 512	512x512x512/temperature.f32
velocity_x	<f4	512 x 512 x 512	512x512x512/velocity_x.f32
velocity_y	<f4	512 x 512 x 512	512x512x512/velocity_y.f32
velocity_z	<f4	512 x 512 x 512	512x512x512/velocity_z.f32

Other files: 512x512x512/template_data.txt

log

sdrbench.dataset("nyx", "log") — folder 512x512x512_log/ — 6 files, 3.2 GB, from SDRBENCH-EXASKY-NYX-512x512x512_log.tar.gz.

The same fields with the two densities log10-transformed (recommended for visualization).

Field	dtype	Shape (C order)	File
baryon_density_log10	<f4	512 x 512 x 512	512x512x512_log/baryon_density_log10.f32
dark_matter_density_log10	<f4	512 x 512 x 512	512x512x512_log/dark_matter_density_log10.f32
temperature	<f4	512 x 512 x 512	512x512x512_log/temperature.f32
velocity_x	<f4	512 x 512 x 512	512x512x512_log/velocity_x.f32
velocity_y	<f4	512 x 512 x 512	512x512x512_log/velocity_y.f32
velocity_z	<f4	512 x 512 x 512	512x512x512_log/velocity_z.f32

Citation

Please cite SDRBench; sdrbench cite nyx prints it:

@inproceedings{zhao2020sdrbench,
  title     = {{SDRBench}: Scientific Data Reduction Benchmark for Lossy Compressors},
  author    = {Zhao, Kai and Di, Sheng and Liang, Xin and Li, Sihuan and Tao, Dingwen and Chen, Zizhong and Cappello, Franck},
  booktitle = {2020 IEEE International Conference on Big Data (Big Data)},
  pages     = {2716--2724},
  year      = {2020}
}

License and terms

The data are distributed by the SDRBench team for research use under the terms of the original data providers listed above (see the SDRBench website); please acknowledge them as requested. Rights remain with the data providers.