SZ4 is under active development and research. Here are some recent developments!
This is the multi-page printable view of this section. Click here to print.
News
- News
- Paper: Accelerating Collective Communications with Lossy Compression on GPU
- Paper: AMRIC: A Novel In Situ Lossy Compression Framework for Efficient I/O in Adaptive Mesh Refinement Applications
- Paper: cuSZp: An Ultra-fast GPU Error-bounded Lossy Compression Framework with Optimized End-to-End Performance
- Paper: ROI preservation
- Paper: wavelet transform bavsed compression
- Paper: TopoSZ: Preserving Topology in Error-Bounded Lossy Compression
- Annoucing the SZ4 Project
- IU News
- FAZ paper receives the best-paper award nomination
- FZ-GPU paper
- QoZ- Quaility-oriented lossy compressor
- David Krasowska
- SZ new version released
- cuSZ new version released
- SZ selected as 2021 R&D100 Awards Winner!!!
- SZ selected as 2021 R&D100 Awards Finalists.
- MMD-SZ released
- vec-SZ released
- cpSZ released
- Roibin-SZ released
- waveSZ released
- QCAT-1.2 released
- cuSZ new version released
- SZauto-1.0.0 released
- cuSZ-0.2.2 released
- SZ3 new version released
- MMD-SZ paper (aka. MDZ) accepted
- QCAT-1.1 released
- cuSZ-0.1.3 released
- SZ-2.1.11 released
- Interp-SZ paper accepted
- kSZ released
- cuSZ-0.1.2 released
- QCAT-1.0 released
- DeepSZ-0.1 released
- cuSZ paper accepted
- SZauto paper accepted
- WaveSZ paper accepted
- Huffman encoding on GPU paper accepted
- cuSZ selected as 3rd Place Winner of ACM Student Research Competition
- DeepSZ paper accepted
- Releases
News
Paper: Accelerating Collective Communications with Lossy Compression on GPU
Jiajun Huang, Sheng Di, Xiaodong Yu, Zizhong Chen, Franck Cappello, Yanfei Guo, and Rajeev Thakur, “Accelerating Collective Communications with Lossy Compression on GPU”, in IEEE/ACM The International Conference for High Performance computing, Networking, Storage and Analysis (IEEE/ACM SC2023), 2023. (1st place of ACM SRC award – Graduates)
Jiajun Huang, Sheng Di, Xiaodong Yu, Zizhong Chen, Franck Cappello, Yanfei Guo, and Rajeev Thakur, "Accelerating Collective Communications with Lossy Compression on GPU", in IEEE/ACM The International Conference for High Performance computing, Networking, Storage and Analysis (IEEE/ACM SC2023), 2023. (1st place of ACM SRC award -- Graduates)
Paper: AMRIC: A Novel In Situ Lossy Compression Framework for Efficient I/O in Adaptive Mesh Refinement Applications
Daoce Wang, Jesus Pulido, Jiannan Tian, Sian Jin, Houjun Tang, Jean Sexton, Sheng Di, Kai Zhao, Bo Fang, Zarija Lukic, Franck Cappello, James Ahrens, Dingwen Tao, “AMRIC: A Novel In Situ Lossy Compression Framework for Efficient I/O in Adaptive Mesh Refinement Applications”, IEEE/ACM The International Conference for High Performance computing, Networking, Storage and Analysis (IEEE/ACM SC2023), 2023.
Daoce Wang, Jesus Pulido, Jiannan Tian, Sian Jin, Houjun Tang, Jean Sexton, Sheng Di, Kai Zhao, Bo Fang, Zarija Lukic, Franck Cappello, James Ahrens, Dingwen Tao, "AMRIC: A Novel In Situ Lossy Compression Framework for Efficient I/O in Adaptive Mesh Refinement Applications", IEEE/ACM The International Conference for High Performance computing, Networking, Storage and Analysis (IEEE/ACM SC2023), 2023.
Paper: cuSZp: An Ultra-fast GPU Error-bounded Lossy Compression Framework with Optimized End-to-End Performance
Yafan Huang, Sheng Di, Xiaodong Yu, Guanpeng Li, Franck Cappello, “cuSZp: An Ultra-fast GPU Error-bounded Lossy Compression Framework with Optimized End-to-End Performance”, in IEEE/ACM The International Conference for High Performance computing, Networking, Storage and Analysis (IEEE/ACM SC2023), 2023.
Yafan Huang, Sheng Di, Xiaodong Yu, Guanpeng Li, Franck Cappello, "cuSZp: An Ultra-fast GPU Error-bounded Lossy Compression Framework with Optimized End-to-End Performance", in IEEE/ACM The International Conference for High Performance computing, Networking, Storage and Analysis (IEEE/ACM SC2023), 2023.
Paper: ROI preservation
Avinash Kethineedi, Jon C. Calhoun, Robert Underwood, Sheng Di, Franck Cappello, “ROI Preservation in Streaming Lossy Compression”, IEEE/ACM The International Conference for High Performance computing, Networking, Storage and Analysis (IEEE/ACM SC2023), 2023
Avinash Kethineedi, Jon C. Calhoun, Robert Underwood, Sheng Di, Franck Cappello, "ROI Preservation in Streaming Lossy Compression", IEEE/ACM The International Conference for High Performance computing, Networking, Storage and Analysis (IEEE/ACM SC2023), 2023
Paper: wavelet transform bavsed compression
Jiajun Huang, Jinyang Liu, Sheng Di, Yujia Zhai, Shixun Wu, Kai Zhao, Zizhong Chen, Yanfei Guo, Franck Cappello, “Exploring Wavelet Transform Usages for Error-bounded Scientific Data Compression”, International Workshop on Big Data Reduction (IEEE IWBDR23) in conjunction with IEEE International Conference on Big Data (IEEE BigData23), 2023
Jiajun Huang, Jinyang Liu, Sheng Di, Yujia Zhai, Shixun Wu, Kai Zhao, Zizhong Chen, Yanfei Guo, Franck Cappello, "Exploring Wavelet Transform Usages for Error-bounded Scientific Data Compression", International Workshop on Big Data Reduction (IEEE IWBDR23) in conjunction with IEEE International Conference on Big Data (IEEE BigData23), 2023
Paper: TopoSZ: Preserving Topology in Error-Bounded Lossy Compression
Researchers at Ohio State have advanced compression for visualization of topologies
abstract …
mention the figure
Annoucing the SZ4 Project
We’ve created the SZ4 project
We are thrilled to announce the launch of the new generation of the SZ lossy compression framework–SZ4, along with its official website! SZ4 is highly modularized, flexible, and easy to customize for specialized use cases, catering to a broad spectrum of scientific applications.
IU News
IU part of $3.2 million grant for cyberinfrastructure research
Indiana University will receive $580,000 of a $3.2 million grant for its collaboration in a National Science Foundation Cyberinfrastructure for Science Innovation project. Dingwen Tao, associate professor of Intelligent Systems Engineering for the Luddy School of Informatics, Computing, and Engineering, is IU’s primary investigator.
The University of Chicago, Ohio State University and the University of Alabama also are collaborating on the project, which aims to develop sustainable computer software to support science research and discovery.
FAZ paper receives the best-paper award nomination
Our [FAZ paper] receives the best-paper award nomination in ICS2023.
QoZ- Quaility-oriented lossy compressor
QoZ: Quaility-oriented lossy compressor is released.
Github link
David Krasowska
David Krasowska won the 1st Place Winnder of the ACM Student Research Competition (Graduate) at SC22.
SZ selected as 2021 R&D100 Awards Winner!!!
SZ has been selected as a winner of 2021 R&D100 Awards!
Link to the R&D website
SZ selected as 2021 R&D100 Awards Finalists.
SZ has been selected as one of 2021 R&D100 Awards Finalists!
Link to the R&D website
Huffman encoding on GPU paper accepted
Our Huffman encoding on GPU paper has been accepted by IPDPS’21.
Link to the paper
cuSZ selected as 3rd Place Winner of ACM Student Research Competition
cuSZ has been selected as the 3rd Place Winner of the ACM Student Research Competition (Graduate) at SC21!
Releases
Foundational Techologies
SZ4 is under active development. We hope to release our first version soon. Here are some of the fundamental technologies that SZ4 will be built on:
SZ3
SZ3 is a modular error-bounded lossy compression framework for scientific datasets.
It provides low-level interfaces and abstractions for writing compressors.
You can find the code for SZ3 on Github, and you can install it with spack spack install sz3
cuSZ
cuSZ is a version of SZ3 optimized for the NVIDIA GPUs and other GPU platforms.
You can find the code for cuSZ on Github
LibPressio
Pressio is latin for compression. LibPressio is a C++ library with C compatible bindings to abstract between different lossless and lossy compressors and their configurations. It solves the problem of having to having to write separate application level code for each lossy compressor that is developed. Instead, users write application level code using LibPressio, and the library will make the correct underlying calls to the compressors. It provides interfaces to represent data, compressors settings, and compressors. Compared to SZ3, it provides much higher level interfaces and abstractions.
You can find LibPressio on GitHub, its documentation on the web, a full tutorial, extensive example codes. You can also install it via spack spack install libpressio