Pith. sign in

REVIEW 1 major objections 4 minor 47 references

Lossless Compression Performance for PETRA III Datasets

T0 review · 1 major / 4 minor · reviewed 2026-08-04 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Lossless compression can shrink PETRA III's non-tape storage by a factor of 1.6–2.1, with the exact ratio set by the allowed throughput.

desk verdict Careful, large-scale compression benchmark with real operational value; the facility-wide extrapolation needs sensitivity analysis before I'd trust the headline ratios. read the letter →

arxiv 2608.00168 v1 pith:HFXYNQCV submitted 2026-07-31 physics.data-an cs.PFphysics.ins-det

Malte Buschmann , Yannis Schumann , Christian Voss , Tigran Mkrtchyan , Philipp Neumann This is my paper · ORCID
classification physics.data-ancs.PFphysics.ins-det
keywords losslesscompressionsynchrotrondataheterogeneityParetofrontZstandardZPAQstoragearchivalPETRAIII
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper is the first systematic measurement of how much lossless compression can shrink data from the PETRA III synchrotron. It gathered more than 212 TiB of raw and processed files from ten experiments across five beamlines, then ran nine general-purpose compressors over a large configuration space to map the trade-off between compression ratio and speed. The headline result: a facility-wide compression ratio of about 2.1 is achievable if throughput is allowed to drop to ~2 MiB/s, and about 1.6 if throughput stays near ~900 MiB/s. The reason this matters is that PETRA III's online storage was growing roughly two-fold every two years, so even a 1.6–2.1 fold reduction could postpone expansion and cut archival cost. Heterogeneous per-category compression outperforms any single uniform policy.

What carries the argument

The carrying mechanism is a Pareto-front-guided local search over compressor configurations, combined with a weighted aggregation step. For each file category (CBF, HDF5/NeXus, Image, Text), the paper benchmarks a 2 GB representative file through more than 10^5 settings, retains configurations within 5% of the two-dimensional Pareto front (ratio vs compression throughput, ratio vs decompression throughput), then re-benchmarks the survivors on 74–208 GB per category. To extrapolate facility-wide, compressed and uncompressed sizes are summed with weights w_i = S_i / s_i proportional to each beamtime's storage share, so the effective ratio is the weighted sum Σ(w_i s_i) divided by Σ(w_i c_i) ra

What would settle it

Benchmark the ~19.5% of PETRA III storage that this study did not cover, or run the same compressors on one new high-throughput detector's raw output; if the unmeasured categories' average ratio is near 1.5 (like HDF5/NeXus) rather than near 5 (like CBF), the extrapolated 2.1 facility-wide ratio will not hold.

Watch

Extended reading notes

Core claim

The central discovery is that PETRA III's storage is not uniformly compressible, and that this heterogeneity, not the compressor choice alone, governs what lossless compression can achieve. Across beamtimes, compression ratios at the same Pareto-optimal settings vary by two orders of magnitude; CBF and Text files compress up to ratios of 5.2 and 7.6, while HDF5/NeXus and Image files top out near 1.7 and 1.5 because they already carry built-in compression. Using weighted aggregation across categories, the paper finds a global trade-off: ZPAQ reaches 2.1 at about 2 MiB/s, while Zstandard in fast mode reaches 1.6 at about 900 MiB/s, with a heterogeneous per-category strategy dominating any sing

Load-bearing premise

The ten sampled beamtimes from five beamlines are representative of all PETRA III online storage, including the 19.5% produced by unmeasured beamlines and all future data.

Editorial extensions

If this is right

  • A heterogeneous compression policy that picks per-category compressors dominates any single uniform compressor across the whole corpus.
  • Adopting a read-heavy archival configuration (Zstandard level ~16) yields a ratio around 1.8 with 971 MiB/s decompression throughput, so one-time compression costs little at retrieval time.
  • CBF and Text files carry most of the exploitable redundancy (max ratios 5.2 and 7.6), so storage savings concentrate there.
  • HDF5/NeXus and Image data are already near their lossless floor, so further general-purpose compression will not shrink them much.
  • Running multiple single-threaded compression instances pinned to separate cache domains scales better than multi-threaded compression for bandwidth-bound high-throughput workloads.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • If the unmeasured beamlines (roughly 19.5% of online storage) turn out to have compressibility closer to HDF5/NeXus than to CBF, the facility-wide 2.1 figure fails; measuring them is the direct test.
  • The byte-offset CBF example suggests detector vendors could co-design entropy coding with the detector-side scheme and recover significant additional storage at acquisition time.
  • The weighted-aggregation method, not the specific numbers, may be the reusable artifact: any facility can run the same Pareto search and weights on its own storage census.
  • With decompression throughput capping around 2–3 GiB/s, interactive retrieval of compressed data may become the bottleneck for users even if archival is fine.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

1 major / 4 minor

Summary. The paper presents a compression benchmark and storage-extrapolation study for PETRA III synchrotron data. The authors assembled a 212.28 TiB corpus from ten beamtimes on five beamlines (p05, p07, p09, p10, p11), manually classified files into five categories, benchmarked nine general-purpose lossless compressors under carefully controlled conditions (NUMA-aware pinning, L3-cache-domain isolation, in-memory I/O, single-threaded measurements), and used a Pareto-guided local search over compressor configurations. They report large between- and within-experiment heterogeneity, find that Zstandard and LZ4 dominate the high-throughput Pareto region while ZPAQ maximizes compression ratio, and claim that heterogeneous per-category compression strategies outperform uniform policies. Extrapolating from the corpus to the full PETRA III non-tape storage system, the paper's headline result is an achievable compression ratio of 2.1 at about 2 MiB/s throughput, or 1.6 at about 900 MiB/s, depending on the throughput budget.

Significance. If the headline extrapolations are robust, the paper provides a quantitative basis for archival storage decisions at a major light source and a template for similar studies at other facilities. The benchmarking methodology is a genuine strength: the authors take reproducibility seriously (public code, containerized experiments, Make-targets), they document hardware and parallelization effects in detail, and the main qualitative conclusions — Zstandard and LZ4 in the high-throughput regime, ZPAQ for maximum ratio, substantial data heterogeneity — are direct measurements rather than fitted model outputs. The weighted aggregation of compressed and uncompressed sizes is conceptually sound, and the quantile bands in Figs. 7–8 honestly display within-directory variability. The main weakness is that the facility-wide ratios in Sec. 5.3 are presented as point estimates without uncertainty bounds or sensitivity analysis, even though they rest on an extrapolation from a small benchmarked sample to the full storage system, including ~19.5% of storage from unmeasured beamlines.

major comments (1)
  1. [Sec. 5.3, Fig. 9; cf. Sec. 3 and Sec. 6] The headline full-system ratios (2.1 at ~2 MiB/s, 1.6 at ~900 MiB/s) are computed by reweighting measured chunk-level compression via P(w_i s_i)/P(w_i c_i), but no uncertainty or sensitivity analysis is reported. Sec. 3 states the five studied beamlines cover only ~80.5% of non-tape storage; the remaining ~19.5% ('others' in Fig. 1) is uncharacterized, and the corpus category mix in Tab. 3 is not verified against that fraction. In addition, each directory contributes at most sixteen 2 GB chunks (Sec. 4.3), while Fig. 7 shows within-category ratios varying by orders of magnitude (e.g., CBF from ~4.3 to >800). The paper's own Sec. 6 caveat that this is a snapshot underscores the need to quantify how much the 1.6/2.1 values could shift; without bootstrap confidence intervals from the chunk-level data or reweighting under alternative assumptions for the unmeasured fraction, the headline numb
minor comments (4)
  1. [Figs. 5–9 captions vs. Sec. 4.2] The figure captions state 'AMD EPYC 7542', while Sec. 4.2 and Table 5 specify 'AMD EPYC 75F3'. These are different processor models. Please correct the inconsistency; reproducibility depends on knowing which CPU was actually used.
  2. [Abstract and Sec. 5.3] The phrase '1.6 at ~900 MiB/s throughput' refers to a single-threaded stream, not facility-wide aggregate throughput. The paper does discuss parallel instances in Sec. 6, but the abstract and the headline sentence in Sec. 5.3 should state this qualification explicitly to avoid over-reading.
  3. [Sec. 4.3] The Pareto search is initialized on a single 2 GB file per category from the largest beamtime. The paper acknowledges this is a compromise, but it would be useful to state in Sec. 5.3 whether the final heterogeneous full-dataset Pareto front changed when the selected configurations were subsequently evaluated on the broader 16-chunk sets. Even a brief qualitative statement would help.
  4. [Code Availability] The text says the code is available via GitLab but gives no URL or repository identifier. A pointer would make the reproducibility claim actionable.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: all central numbers are direct measurements or weighted re-aggregations of measured sizes; the only soft point is an acknowledged representativeness extrapolation, which is not circular.

full rationale

The paper's headline claims (2.1 at ~2 MiB/s, 1.6 at ~900 MiB/s) are computed by taking directly measured compressed and uncompressed sizes from benchmarked files and aggregating them with storage-volume weights: "we aggregate compressed and uncompressed file sizes separately and compute the effective compression ratio from their weighted sums, P(w_i s_i)/P(w_i c_i)." No target quantity is defined in terms of a fitted parameter, and no conclusion is assumed in its own derivation. The Pareto-guided local neighborhood search is configuration selection over compressor settings, not a fit of the headline result; all final ratios come from independent compression runs on held-out 2 GB chunks per subdirectory/beamtime. The extrapolation from ten beamtimes to the full PETRA III non-tape storage relies on the assumption that the measured beamlines (80.5% of storage) are representative of the remaining 19.5% and of future data, and the paper itself acknowledges that "the present work necessarily represents a snapshot in time." That is an external-validity limitation, not circularity: it does not make the benchmarked numbers equivalent to their inputs by construction. There are no load-bearing self-citations, no imported uniqueness theorems, and no ansatz smuggled in via citation. The compression ratios are measurements of external corpus content, and the aggregation formula is a straightforward weighted sum of observable quantities. Therefore the circularity score is 0.

Assumptions & free parameters 5 free parameters · 4 assumptions · 0 invented entities

The paper is an empirical benchmark; it introduces no fitted scientific model. The headline numbers are weighted re-aggregations of direct measurements. The ledger therefore consists of representative-sampling and hardware assumptions plus hand-chosen benchmark thresholds, rather than fitted parameters or invented entities.

free parameters (5)
  • Near-Pareto 5% threshold = 5%
    Configurations within 5% of the current Pareto front are explored in the local search (Sec. 4.3); chosen by hand and influences which compressors appear Pareto-optimal.
  • Benchmark payload size = 2 GB
    A single 2 GB tar per category is used in the optimization stage and 2 GB chunks in validation (Sec. 4.2); a compromise between representativeness and runtime.
  • Throughput timeout = 1 MiB/s
    Tasks slower than 1 MiB/s are timed out, which excludes extremely high-ratio configurations from the reported fronts (Sec. 4.2).
  • Optimization-stage representative file choice = largest beamtime per category
    For each category, the initial configuration search uses one 2 GB file from the beamtime contributing the largest storage volume (Sec. 4.3), which can bias selected configurations toward that beamtime.
  • Validation chunk count = 16 chunks of 2 GB
    Selected configurations are re-evaluated on up to sixteen 2 GB chunks per subdirectory/beamtime when at least 20 GB is available (Sec. 4.3); this sampling threshold affects the statistical weight and uncertainty of the final aggregates.
assumptions (4)
  • domain assumption The five measured beamlines (p05, p07, p09, p10, p11) are representative of full PETRA III non-tape storage, including the ~19.5% from other beamlines.
    Sec. 3 notes the measured beamlines account for ~80.5% of stored data; Sec. 5.3 extrapolates to the full storage system without reweighting or validating the remaining 19.5%.
  • domain assumption Compression performance on the selected 2 GB files/chunks represents each file category across beamtimes.
    Sec. 4.3 explicitly acknowledges a single benchmark file may not perfectly represent category statistics, then assumes relative comparisons hold; the final aggregate ratios depend on this assumption.
  • domain assumption Manual suffix-to-category assignment is correct.
    Sec. 3 states each unique suffix was manually attributed to one of five categories; misclassification would shift category-level compression ratios.
  • domain assumption Benchmark hardware and single-thread execution reflect production archival environments.
    Sec. 4.2 uses dual-socket AMD EPYC 75F3 nodes; throughput numbers in Tab. 5 are quoted for this hardware and may differ on production hardware.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Lossless Compression Performance for PETRA III Datasets." pith.science (2026). https://pith.science/paper/HFXYNQCV

@misc{pith2026260800168,
  author       = {Pith},
  title        = {Pith review of: Lossless Compression Performance for PETRA III Datasets},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/HFXYNQCV}},
  note         = {Machine review of arXiv:2608.00168}
}
abstract

Large-scale research facilities increasingly face the challenge of managing rapidly growing data volumes while maintaining sustainable archival infrastructures. We present the first comprehensive study of data heterogeneity and lossless general-purpose compression performance for representative datasets from the PETRA III synchrotron radiation source. Our corpus comprises more than 212 TiB of raw and processed data from ten experiments spanning multiple beamlines, detector systems, and scientific workflows. We observe substantial heterogeneity both between and within experiments, resulting in compression ratios that vary by more than two orders of magnitude across datasets. Evaluating nine widely used lossless compression tools, we find that Zstandard and LZ4 consistently occupy the high-throughput region of the Pareto front, whereas ZPAQ achieves the highest compression ratios. Furthermore, heterogeneous compression strategies that adapt compressor choice to the underlying file category outperform uniform compression policies. Extrapolating from the benchmarked datasets to the full PETRA III non-tape storage system, we estimate achievable compression ratios ranging from approximately 1.6 at $\sim$900 MiB/s throughput to 2.1 at $\sim$2 MiB/s. These results provide a quantitative basis for future archival and storage decisions at PETRA III, its future successor, PETRA IV, and other large-scale scientific facilities.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

47 extracted references · 15 canonical work pages

  1. [1]

    Wilkinson, Mark D. and Dumontier, Michel and Aalbersberg, IJsbrand Jan and Appleton, Gabrielle and Axton, Myles and Baak, Arie and Blomberg, Niklas and Boiten, Jan-Willem and da Silva Santos, Luiz Bonino and Bourne, Philip E. and Bouwman, Jildau and Brookes, Anthony J. and Clark, Tim and Crosas, Mercè and Dillo, Ingrid and Dumon, Olivier and Edmunds, Scot...

    2016

  2. [2]

    Time-resolved crystallography of boric acid binding to the active site serine of the beta-lactamase CTX-M-14 and subsequent 1,2-diol esterification , year =

    Prester, Andreas and Perbandt, Markus and Galchenkova, Marina and Oberthuer, Dominik and Werner, Nadine and Henkel, Alessandra and Maracke, Julia and Yefanov, Oleksandr and Hakanpää, Johanna and Pompidor, Guillaume and Meyer, Jan and Chapman, Henry and Aepfelbacher, Martin and Hinrichs, Winfried and Rohde, Holger and Betzel, Christian , journal =. Time-re...

  3. [3]

    Günther, Sebastian and Reinke, Patrick Y. A. and Fernández-García, Yaiza and Lieske, Julia and Lane, Thomas J. and Ginn, Helen M. and Koua, Faisal H. M. and Ehrt, Christiane and Ewert, Wiebke and Oberthuer, Dominik and Yefanov, Oleksandr and Meier, Susanne and Lorenzen, Kristina and Krichel, Boris and Kopicki, Janine-Denise and Gelisio, Luca and Brehm, Wo...

  4. [4]

    Stochastic atomic acceleration during the X-ray-induced fluidization of a silica glass , year =

    Dallari, Francesco and Martinelli, Alessandro and Caporaletti, Federico and Sprung, Michael and Baldi, Giacomo and Monaco, Giulio , journal =. Stochastic atomic acceleration during the X-ray-induced fluidization of a silica glass , year =. doi:10.1073/pnas.2213182120 , publisher =

    work page doi:10.1073/pnas.2213182120

  5. [5]

    Reaching the Yield Point of a Glass During X-Ray Irradiation , year =

    Martinelli, Alessandro and Caporaletti, Federico and Dallari, Francesco and Sprung, Michael and Westermeier, Fabian and Baldi, Giacomo and Monaco, Giulio , journal =. Reaching the Yield Point of a Glass During X-Ray Irradiation , year =. doi:10.1103/physrevx.13.041031 , publisher =

    work page doi:10.1103/physrevx.13.041031

  6. [6]

    and Richter, Adrian K

    Boudinot, Brendon E. and Richter, Adrian K. and Hammel, Jörg U. and Szwedo, Jacek and Bojarski, Błażej and Perrichot, Vincent , journal =. Genomic-Phenomic Reciprocal Illumination: Desyopone hereon gen. et sp. nov., an Exceptional Aneuretine-like Fossil Ant from Ethiopian Amber (Hymenoptera: Formicidae: Ponerinae) , year =. doi:10.3390/insects13090796 , p...

    work page doi:10.3390/insects13090796

  7. [7]

    and Feldbach, E

    Museur, L. and Feldbach, E. and Kotlov, A. and Kitaura, M. and Kanaev, A. , journal =. Donor-acceptor pair transitions in MgAl2O4 spinel , year =. doi:10.1016/j.jlumin.2023.120235 , publisher =

    arXiv 2023

  8. [8]

    and Leupold, O

    Franz, H. and Leupold, O. and Röhlsberger, R. and Roth, S.V. and Seeck, O.H. and Spengler, J. and Strempfer, J. and Tischer, M. and Viefhaus, J. and Weckert, E. and Wroblewski, T. , journal =. Technical Report: PETRA III: DESY’s New High Brilliance Third Generation Synchrotron Radiation Source , year =. doi:10.1080/08940880601064984 , publisher =

    work page doi:10.1080/08940880601064984

Show all 47 references
  1. [9]

    The ESRF: Four Years of Extremely Brilliant Science , year =

    Cloetens, Peter and Daillant, Jean and Detlefs, Carsten and Drnec, Jakub and Frey, Johannes and Hernandez, Jean-Alexis and Honkimaki, Veijo and Krisch, Michael and Martínez-Criado, Gema and McCarthy, Joanne and Mezouar, Mohamed and Di Michiel, Marco and Mitchell, Edward and Pa...

    2025

  2. [10]

    Schroer, Christian G. and Agapov, Ilya and Brefeld, Werner and Brinkmann, Reinhard and Chae, Yong-Chul and Chao, Hung-Chun and Eriksson, Mikael and Keil, Joachim and Nuel Gavaldà, Xavier and Röhlsberger, Ralf and Seeck, Oliver H. and Sprung, Michael and Tischer, Markus and Wan...

    doi

  3. [11]

    High Performance Parallel I/o , year =

    Prabhat and Koziol, Quincey , publisher =. High Performance Parallel I/o , year =

  4. [12]

    BLISS - Experiments Control for ESRF EBS Beamlines , year =

    Guijarro, Matias and Beteva, Antonia and Coutinho, Tiago and Dominguez, Marie-Christine and Guilloud, Cyril and Homs, Alejandro and Meyer, Jens and Michel, Vincent and Papillon, Emmanuel and Perez, Manuel and Petitdemange, SéBastien , journal =. BLISS - Experiments Control fo...

    doi

  5. [13]

    Tonellot and Zizhong Chen and Franck Capello , publisher =

    Kai Zhao and Sheng Di and Maxim Dmitriev and Thierry-Laurent D. Tonellot and Zizhong Chen and Franck Capello , publisher =. Optimizing Error-Bounded Lossy Compression for Scientific Data by Dynamic Spline Interpolation , year =

  6. [14]

    Fixed-Rate Compressed Floating-Point Arrays

    Lindstrom, Peter , journal =. Fixed-Rate Compressed Floating-Point Arrays. , year =. doi:10.1109/TVCG.2014.2346458 , issn-linking =

    2014

  7. [15]

    MGARD: A multigrid framework for high-performance, error-controlled data compression and refactoring , year =

    Qian Gong and Jieyang Chen and Ben Whitney and Xin Liang and Viktor Reshniak and Tania Banerjee and Jaemoon Lee and Anand Rangarajan and Lipeng Wan and Nicolas Vidal and Qing Liu and Ana Gainaru and Norbert Podhorszki and Richard Archibald and Sanjay Ranka and Scott Klasky , j...

    2023

  8. [16]

    Fast and efficient compression of floating-point data

    Lindstrom, Peter and Isenburg, Martin , journal =. Fast and efficient compression of floating-point data. , year =. doi:10.1109/TVCG.2006.143 , issn-linking =

    2006 doi

  9. [17]

    2021 , abstract =

    Yann Collet and Murray Kucherawy , howpublished =. 2021 , abstract =. doi:10.17487/RFC8878 , number =

    2021 doi

  10. [18]

    Brotli: A General-Purpose Data Compressor , year =

    Alakuijala, Jyrki and Farruggia, Andrea and Ferragina, Paolo and Kliuchnikov, Eugene and Obryk, Robert and Szabadka, Zoltan and Vandevenne, Lode , journal =. Brotli: A General-Purpose Data Compressor , year =. doi:10.1145/3231935 , issue_date =

    doi

  11. [19]

    Peter Deutsch , howpublished =

    L. Peter Deutsch , howpublished =. 1996 , abstract =. doi:10.17487/RFC1951 , number =

    1996 doi

  12. [20]

    Peter Deutsch , title =

    L. Peter Deutsch , title =. RFC 1952 , year =

    1952

  13. [21]

    and Bernstein, Herbert J

    Könnecke, Mark and Akeroyd, Frederick A. and Bernstein, Herbert J. and Brewster, Aaron S. and Campbell, Stuart I. and Clausen, Björn and Cottrell, Stephen and Hoffmann, Jens Uwe and Jemian, Pete R. and Männicke, David and Osborn, Raymond and Peterson, Peter F. and Richter, Tob...

    doi

  14. [22]

    Bernstein, H. J. and Hammersley, A. P. , pages =. Specification of the Crystallographic Binary File (CBF/imgCIF) , year =. International Tables for Crystallography , doi =

  15. [23]

    and Amiri, M

    Masui, K. and Amiri, M. and Connor, L. and Deng, M. and Fandino, M. and Höfer, C. and Halpern, M. and Hanna, D. and Hincks, A.D. and Hinshaw, G. and Parra, J.M. and Newburgh, L.B. and Shaw, J.R. and Vanderlinde, K. , journal =. A compression scheme for radio data in high perfo...

    2015 doi

  16. [24]

    Hierarchical Storage Support and Management for Large-Scale Multidimensional Array Database Management Systems , year =

    Reiner, Bernd and Hahn, Karl and Höfling, Gabriele and Baumann, Peter , pages =. Hierarchical Storage Support and Management for Large-Scale Multidimensional Array Database Management Systems , year =. Database and Expert Systems Applications , doi =

  17. [25]

    , booktitle =

    Brown, Paul G. , booktitle =. Overview of sciDB: large scale array storage, processing and analysis , year =

  18. [26]

    Fast lossless images compression for synchrotron radiation facility using deep learning and hybrid architecture , year =

    Min-xing, Zhang and Shi-yuan, Fu and Yu, Gao and Yao-dong, Cheng and Mustofa, Abdulhafiz Ahmed and Gang, Chen , journal =. Fast lossless images compression for synchrotron radiation facility using deep learning and hybrid architecture , year =. doi:10.1007/s41605-024-00490-9 ,...

    doi

  19. [27]

    and Soares, Alexei S

    Bernstein, Herbert J. and Soares, Alexei S. and Horvat, Kimberly and Jakoncic, Jean , journal =. Massive compression for high data rate macromolecular crystallography (HDRMX): impact on diffraction data and subsequent structural analysis , year =. doi:10.1107/s1600577525000396...

    doi

  20. [28]

    Modern lossless compression techniques: Review, comparison and analysis , year =

    Gupta, Apoorv and Bansal, Aman and Khanduja, Vidhi , booktitle =. Modern lossless compression techniques: Review, comparison and analysis , year =

  21. [29]

    Comparison of Lossless Data Compression Techniques , year =

    Gopinath, Athira and Ravisankar, M , booktitle =. Comparison of Lossless Data Compression Techniques , year =

  22. [30]

    A survey and benchmark evaluation for neural-network-based lossless universal compressors toward multi-source data , year =

    Sun, Hui and Ma, Huidong and Ling, Feng and Xie, Haonan and Sun, Yongxia and Yi, Liping and Yan, Meng and Zhong, Cheng and Liu, Xiaoguang and Wang, Gang , journal =. A survey and benchmark evaluation for neural-network-based lossless universal compressors toward multi-source d...

    doi

  23. [31]

    Chen, Xinyu and Tian, Jiannan and Beaver, Ian and Freeman, Cynthia and Yan, Yan and Wang, Jianguo and Tao, Dingwen , title =. Proc. VLDB Endow. , month = feb, pages =. 2024 , issue_date =. doi:10.14778/3648160.3648180 , abstract =

    2024

  24. [32]

    and Barty, Anton and Chapman, Henry N

    Galchenkova, Marina and Tolstikova, Alexandra and Klopprogge, Bjarne and Sprenger, Janina and Oberthuer, Dominik and Brehm, Wolfgang and White, Thomas A. and Barty, Anton and Chapman, Henry N. and Yefanov, Oleksandr , journal =. Data reduction in protein serial crystallography...

    doi

  25. [33]

    Intelligent compression for synchrotron radiation source image , year =

    Fu, Shiyuan and Wang, Lu and Cheng, Yaodong and Chen, Gang , journal =. Intelligent compression for synchrotron radiation source image , year =. doi:10.1051/epjconf/202125103073 , editor =

  26. [34]

    FPGA Acceleration of Zstd Compression Algorithm , year =

    Chen, Jianyu and Daverveldt, Maurice and Al-Ars, Zaid , booktitle =. FPGA Acceleration of Zstd Compression Algorithm , year =

  27. [35]

    LZ4 compression algorithm on FPGA , year =

    Bartik, Matej and Ubik, Sven and Kubalik, Pavel , booktitle =. LZ4 compression algorithm on FPGA , year =

  28. [36]

    doi:10.5281/zenodo.17808614 , url =

    2025 , publisher =. doi:10.5281/zenodo.17808614 , url =

    2025 doi

  29. [37]

    EIGER detectors in biological crystallography , year =

    Andreas Förster and Stefan Brandstetter and Marcus Mueller and Clemens Schulze-Briese , institution =. EIGER detectors in biological crystallography , year =

  30. [38]

    2008 , howpublished =

    Con Kolivas , title =. 2008 , howpublished =

    2008

  31. [39]

    1996 , howpublished =

    Markus Franz Xaver Johannes Oberhumer , title =. 1996 , howpublished =

    1996

  32. [40]

    2011 , howpublished =

    Yann Collet , title =. 2011 , howpublished =

    2011

  33. [41]

    2009 , howpublished =

    Lasse Collin , title =. 2009 , howpublished =

    2009

  34. [42]

    2011 , howpublished =

    Matt Mahoney , title =. 2011 , howpublished =

    2011

  35. [43]

    1999 , howpublished =

    Igor Pavlov , title =. 1999 , howpublished =

    1999

  36. [44]

    1998 , howpublished =

    Julian Seward , title =. 1998 , howpublished =

    1998

  37. [45]

    , booktitle=

    Shkarin, D. , booktitle=. PPM: one step to practicality , year=

  38. [46]

    2023 , Journal =

    The European strategy for accelerator-based photon science , Author =. 2023 , Journal =. doi:10.1140/epjp/s13360-023-03947-w , Number =

    2023 doi

  39. [47]

    Dolcet and J.-D

    P. Dolcet and J.-D. Grunwaldt and S. Hövelmann and L. Amelung and A. Schneidewind , title =. Synchrotron Radiation News , volume =. 2023 , publisher =. doi:10.1080/08940886.2023.2277655 , URL =

    2023

Pith tools

Reviewed August 4, 2026 · model on record in the stance chip above.