English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

The fast growth of computational power and scales of modern super-computingsystems have raised great challenges for the management of exascale scientificdata. To maintain the usability of scientific data, error-bound lossycompression is proposed and developed as an essential technique for the sizereduction of scientific data with constrained data distortion. Among thediverse datasets generated by various scientific simulations, certain datasetscannot be effectively compressed by existing error-bounded lossy compressorswith traditional techniques. The recent success of Artificial Intelligence hasinspired several researchers to integrate neural networks into error-boundedlossy compressors. However, those works still suffer from limited compressionratios and/or extremely low efficiencies. To address those issues and improvethe compression on the hard-to-compress datasets, in this paper, we proposeSRN-SZ, which is a deep learning-based scientific error-bounded lossycompressor leveraging the hierarchical data grid expansion paradigm implementedby super-resolution neural networks. SRN-SZ applies the most advancedsuper-resolution network HAT for its compression, which is free of time-costingper-data training. In experiments compared with various state-of-the-artcompressors, SRN-SZ achieves up to 75% compression ratio improvements under thesame error bound and up to 80% compression ratio improvements under the samePSNR than the second-best compressor.

SRN-SZ: Deep Leaning-Based Scientific Error-bounded Lossy Compression with Super-resolution Neural Networks