cuSZ-$i$: High-Ratio Scientific Lossy Compression on GPUs with Optimized Multi-Level Interpolation

TL;DR

cuSZ-$i$ is a GPU-based scientific lossy compressor that significantly improves compression ratios and data quality through novel interpolation prediction, optimized Huffman encoding, and integration of NVIDIA Bitcomp, outperforming existing GPU compressors.

Contribution

The paper introduces cuSZ-$i$, a novel GPU-based lossy compressor with advanced prediction, encoding, and ratio-enhancement techniques, achieving substantially higher compression ratios.

Findings

cuSZ-$i$ achieves 476% higher compression ratio than the second-best GPU compressor.

It significantly improves data quality and compression efficiency.

cuSZ-$i$ outperforms existing compressors in real-world HPC applications.

Abstract

Error-bounded lossy compression is a critical technique for significantly reducing scientific data volumes. Compared to CPU-based compressors, GPU-based compressors exhibit substantially higher throughputs, fitting better for today's HPC applications. However, the critical limitations of existing GPU-based compressors are their low compression ratios and qualities, severely restricting their applicability. To overcome these, we introduce a new GPU-based error-bounded scientific lossy compressor named cuSZ-$i$, with the following contributions: (1) A novel GPU-optimized interpolation-based prediction method significantly improves the compression ratio and decompression data quality. (2) The Huffman encoding module in cuSZ-$i$ is optimized for better efficiency. (3) cuSZ-$i$ is the first to integrate the NVIDIA Bitcomp-lossless as an additional compression-ratio-enhancing module. Evaluations show that cuSZ-$i$ significantly outperforms other latest GPU-based lossy compressors in compression ratio under the same error bound (hence, the desired quality), showcasing a 476% advantage over the second-best. This leads to cuSZ-$i$'s optimized performance in several real-world use cases.