Boosting Scientific Error-Bounded Lossy Compression through Optimized Synergistic Lossy-Lossless Orchestration

TL;DR

cuSZ-Hi is a GPU-based scientific lossy compressor that significantly improves compression ratios and throughput by optimizing data prediction and lossless encoding techniques, suitable for high-throughput scientific workflows.

Contribution

The paper introduces cuSZ-Hi, a novel GPU-optimized framework that enhances scientific data compression through adaptive prediction and tailored lossless encoding pipelines.

Findings

Achieves up to 249% higher compression ratio than existing methods.

Maintains comparable or better throughput on GPU architectures.

Demonstrates significant improvements on benchmark datasets.

Abstract

As high-performance computing architectures evolve, more scientific computing workflows are being deployed on advanced computing platforms such as GPUs. These workflows can produce raw data at extremely high throughputs, requiring urgent high-ratio and low-latency error-bounded data compression solutions. In this paper, we propose cuSZ-Hi, an optimized high-ratio GPU-based scientific error-bounded lossy compressor with a flexible, domain-irrelevant, and fully open-source framework design. Our novel contributions are: 1) We maximally optimize the parallelized interpolation-based data prediction scheme on GPUs, enabling the full functionalities of interpolation-based scientific data prediction that are adaptive to diverse data characteristics; 2) We thoroughly explore and investigate lossless data encoding techniques, then craft and incorporate the best-fit lossless encoding pipelines for maximizing the compression ratio of cuSZ-Hi; 3) We systematically evaluate cuSZ-Hi on benchmarking datasets together with representative baselines. Compared to existing state-of-the-art scientific lossy compressors, with comparative or better throughput than existing high-ratio scientific error-bounded lossy compressors on GPUs, cuSZ-Hi can achieve up to 249% compression ratio improvement under the same error bound, and up to 215% compression ratio improvement under the same decompression data PSNR.