SIMD Lossy Compression for Scientific Data

TL;DR

This paper presents a SIMD vectorized CPU implementation of the SZ lossy compressor for scientific data, achieving significant speedups and improved compression quality by optimizing parallel prediction/quantization techniques.

Contribution

It introduces a SIMD CPU version of SZ with heuristics for block size and vector length, and explores padding strategies to enhance parallelism and reduce outliers.

Findings

Up to 32% better rate-distortion performance.

Up to 15x faster than SZ-1.4.

Padding reduces outliers by 100% in some cases.

Abstract

Modern HPC applications produce increasingly large amounts of data, which limits the performance of current extreme-scale systems. Data reduction techniques, such as lossy compression, help to mitigate this issue by decreasing the size of data generated by these applications. SZ, a current state-of-the-art lossy compressor, is able to achieve high compression ratios, but the prediction/quantization methods used introduce dependencies which prevent parallelizing this step of the compression. Recent work proposes a parallel dual prediction/quantization algorithm for GPUs which removes these dependencies. However, some HPC systems and applications do not use GPUs, and could still benefit from the fine-grained parallelism of this method. Using the dual-quantization technique, we implement and optimize a SIMD vectorized CPU version of SZ, and create a heuristic for selecting the optimal block size and vector length. We also investigate the effect of non-zero block padding values to decrease the number of unpredictable values along compression block borders. We measure performance of vecSZ against an O3 optimized CPU version of SZ using dual-quantization, pSZ, as well as SZ-1.4. We evaluate our vectorized version, vecSZ, on the Intel Skylake and AMD Rome architectures using real-world scientific datasets. We find that applying alternative padding reduces the number of outliers by 100\% for some configurations. Our implementation also results in up to 32\% improvement in rate-distortion and up to 15$\times$ speedup over SZ-1.4, achieving a prediction and quantization bandwidth in excess of 3.4 GB/s.