Home - News ( United States | United Kingdom | Italy | Germany ) - Football scores

Showing content from https://github.com/szcompressor/SZ3/tree/QoZ below:

GitHub - szcompressor/SZ3 at QoZ

(C) 2016 by Mathematics and Computer Science (MCS), Argonne National Laboratory. See COPYRIGHT in top-level directory.

• Major Authors: Sheng Di, Kai Zhao, Xin Liang, Jinyang Liu
• Supervisor: Franck Cappello
• Other Contributors: Robert Underwood, Sihuan Li, Ali M. Gok

This branch of SZ3 integrates QoZ, the quality-oriented compression optimization module. In the sz3 executable, you can introduce new arguments to test QoZ-integrated SZ3 compression:

• -q 0: SZ3.1 compression in QoZ backend.
• -q 1: QoZ 1.1 compression.
• -q 2 or -q 3: 2 intermediate optimization levels of QoZ 2.0 compression (having faster speeds but slightly worse rate-distortion).
• -q 4: Full QoZ 2.0 compression (for the results reported in the paper).
• -l: Activate Lorenzo tuning in QoZ compression.
• -V: Show auto-tuning information during QoZ compression.
• -T CR: Compression-ratio-preferred auto-tuning (the default is rate-PSNR-preferred -T PSNR).

You can also add the following parameters to the config file instead: QoZ (-q), testLorenzo (-l), verbose (-V), and tuningTarget (-T). Without new arguments/parameters, the original sz3 will be executed.

Notice: Not setting -q or setting -q -1 will cover all other new settings, forcing the usage of the original SZ3. You need to set -q to at least 0 to activate QoZ features.

If you used QoZ in your work, the appropriate citations are:

• QoZ 2.0 (HPEZ): [SIGMOD 24] Jinyang Liu, Sheng Di, Kai Zhao, Xin Liang, Sian Jin, Zizhe Jian, Jiajun Huang, Shixun Wu, Zizhong Chen, and Franck Cappello. 2023. "High-performance Effective Scientific Error-bounded Lossy Compression with Auto-tuned Multi-component Interpolation." in Proceedings of the ACM on Management of Data 2, no. 1 (2024): 1-27.
• QoZ 1.0: [SC 22] Jinyang Liu, Sheng Di, Sian Jin, Kai Zhao, Xin Liang, Zizhong Chen, and Franck Cappello. "Dynamic quality metric oriented error bounded lossy compression for scientific datasets." In SC22: International Conference for High Performance Computing, Networking, Storage and Analysis, pp. 1-15. IEEE, 2022.

• mkdir build && cd build
• cmake -DCMAKE_INSTALL_PREFIX:PATH=[INSTALL_DIR] ..
• make
• make install

Then, you'll find all the executables in [INSTALL_DIR]/bin and header files in [INSTALL_DIR]/include

• Zstandard (https://facebook.github.io/zstd/). Zstandard v1.4.5 is included and will be used if libzstd can not be found by pkg-config.

You can use the executable 'sz3' command to do the compression/decompression.

SZ3 simplifies command line arguments in the previous version. If you are a new user, please follow the instructions given by the executable.

For backward compatibility, most of the SZ2 command line parameters are supported in SZ3. Exceptions are listed below. Scripts without parameters below should work fine by replacing SZ2 with SZ3.

• Located in 'include/SZ3/api/sz.hpp'.
• Requiring a modern C++ compiler.
• Different with SZ2 API.

• Located in 'tools/sz3c/include/sz3c.h'
• Compatible with SZ2 API

• Located in 'tools/pysz/pysz.py'
• Test file provided ('tools/pysz/test.py')
• Compatible with both SZ3 and SZ2
• Requiring SZ2/3 dynamic library

• Special thanks to Oscar Mojica for providing the Fortran API
• Visit this Github repository for details

• Located in 'tools/H5Z-SZ3'
• Please add "-DBUILD_H5Z_FILTER=ON" to enable this function for CMake.
• sz3ToHDF5 and HDF5ToSz3 are provided for testing.

Version New features

• SZ 3.0.0 SZ3 is the C++ version of SZ with modular and composable design.
• SZ 3.0.1 Improve the build process.
• SZ 3.1.0 The default algorithm is now interpolation+Lorenzo.
• SZ 3.1.1 Add OpenMP support. Works for all algorithms. Please enable it using the config file.
• SZ 3.1.2 Support configuration file (INI format). Example can be found in 'tools/sz3/sz3.config'.
• SZ 3.1.3 Support more error control mode: PSNR, L2Norm, ABS_AND_REL, ABS_OR_REL. Support INT32 and INT64 datatype.
• SZ 3.1.4 Support running on Windows. Please refer to #5 (comment) for instructions.
• SZ 3.1.5 Support HDF5 by H5Z-SZ3. Please add "-DBUILD_H5Z_FILTER=ON" to enable this function for CMake.
• SZ 3.1.6 Support C API and Python API.
• SZ 3.1.7 Initial MDZ(https://github.com/szcompressor/SZ3/tree/master/tools/mdz) support.
• SZ 3.1.8 namespace changed from SZ to SZ3. H5Z-SZ3 supports configuration file now.
• SZ 3.2.0 API reconstructed for FZ. H5Z-SZ3 rewrite. Compression version checking.

Kindly note: If you mention SZ in your paper, the most appropriate citation is including these three references (TBD22, ICDE21, Bigdata18), because they cover the design and implementation of the latest version of SZ.

• SZ3 Framework: Xin Liang, Kai Zhao, Sheng Di, Sihuan Li, Robert Underwood, Ali M Gok, Jiannan Tian, Junjing Deng, Jon C Calhoun, Dingwen Tao, Zizhong Chen, and Franck Cappello. "SZ3: A modular framework for composing prediction-based error-bounded lossy compressors", IEEE Transactions on Big Data (TBD 22).

• SZ3 Algorithm: Kai Zhao, Sheng Di, Maxim Dmitriev, Thierry-Laurent D. Tonellot, Zizhong Chen, and Franck Cappello. "Optimizing Error-Bounded Lossy Compression for Scientific Data by Dynamic Spline Interpolation" , Proceeding of the 37th IEEE International Conference on Data Engineering (ICDE 21), Chania, Crete, Greece, Apr 19 - 22, 2021.

• SZauto: Kai Zhao, Sheng Di, Xin Liang, Sihuan Li, Dingwen Tao, Zizhong Chen, and Franck Cappello. "Significantly Improving Lossy Compression for HPC Datasets with Second-Order Prediction and Parameter Optimization" , Proceedings of the 29th International Symposium on High-Performance Parallel and Distributed Computing (HPDC 20), Stockholm, Sweden, 2020. (code: https://github.com/szcompressor/SZauto/)

• SZ 2.0+: Xin Liang, Sheng Di, Dingwen Tao, Zizhong Chen, Franck Cappello, "Error-Controlled Lossy Compression Optimized for High Compression Ratios of Scientific Datasets" , in IEEE International Conference on Big Data (Bigdata 2018), Seattle, WA, USA, 2018.

• SZ 1.4.0-1.4.13: Dingwen Tao, Sheng Di, Franck Cappello. "Significantly Improving Lossy Compression for Scientific Data Sets Based on Multidimensional Prediction and Error-Controlled Quantization" , in IEEE International Parallel and Distributed Processing Symposium (IPDPS 2017), Orlando, Florida, USA, 2017.

• SZ 0.1-1.0: Sheng Di, Franck Cappello. "Fast Error-bounded Lossy HPC Data Compression with SZ", in IEEE International Parallel and Distributed Processing Symposium (IPDPS 2016), Chicago, IL, USA, 2016.

• Point-wise relative error bound mode (i.e., PW_REL): Xin Liang, Sheng Di, Dingwen Tao, Zizhong Chen, Franck Cappello, "An Efficient Transformation Scheme for Lossy Data Compression with Point-wise Relative Error Bound" , in IEEE International Conference on Clustering Computing (CLUSTER 2018), Belfast, UK, 2018. (Best Paper)

RetroSearch is an open source project built by @garambo | Open a GitHub Issue

Search and Browse the WWW like it's 1997 | Search results from DuckDuckGo

HTML: 3.2 | Encoding: UTF-8 | Version: 0.7.4