README.md

metadata

license: other
license_name: sdrbench
license_link: https://sdrbench.github.io/
pretty_name: 'SDRBench: QMCPACK'
tags:
  - scientific-data
  - lossy-compression
  - sdrbench
  - hpc
  - sz3
configs:
  - config_name: files
    data_files: files.jsonl
    default: true

SDRBench — QMCPACK

Many-body ab initio Quantum Monte Carlo (QMCPACK): 288 einspline orbitals on a 115 x 69 x 69 grid, single precision, little-endian.

Raw binary arrays, not an Arrow/Parquet dataset. datasets.load_dataset("sdrbench/qmcpack") 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 QMCPACK 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 1518533bae). metadata/ holds the original SDRBench property and template files. Kept in sync automatically by szcompressor/sdrbench.

• Data provider: QMCPACK performance test; contact Ye Luo (yeluo@anl.gov)
• Size: 630.7 MB 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("qmcpack")                # default variant "preconditioned"
print(ds.fields)
x = ds["einspline"]                     # numpy array, dtype <f4, shape (288, 115, 69, 69)

# 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("qmcpack", 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/qmcpack", "115x69x69x288/einspline_115_69_69_288.f32", repo_type="dataset", revision="1518533baeeffa9dad33910ff639029bfac91f88")
x = np.fromfile(p, dtype="<f4").reshape((115, 69, 69, 288)).transpose(3, 0, 1, 2)  # -> (288, 115, 69, 69)

Variants and fields

preconditioned (default)

sdrbench.dataset("qmcpack", "preconditioned") — derived from 115x69x69x288/, no extra files.

(orbital, x, y, z) = (288, 115, 69, 69): each orbital is a contiguous, smooth 3D block. This is the layout of the SDRBench command examples (einspline_288_115_69_69.pre.f32); computed on load from the stored file.

Field	dtype	Shape (C order)	File
einspline	<f4	288 x 115 x 69 x 69	115x69x69x288/einspline_115_69_69_288.f32

original

sdrbench.dataset("qmcpack", "original") — folder 115x69x69x288/ — 3 files, 630.7 MB, from SDRBENCH-QMCPack.tar.gz.

The layout QMCPACK uses in memory: (x, y, z, orbital) = (115, 69, 69, 288), orbital index fastest.

Field	dtype	Shape (C order)	File
einspline	<f4	115 x 69 x 69 x 288	115x69x69x288/einspline_115_69_69_288.f32

Other files: 115x69x69x288/SDRBENCH-QMCPACK-115x69x69x288-property.txt, 115x69x69x288/template_data.txt

Citation

Please cite SDRBench; sdrbench cite qmcpack 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.