Datasets for Benchmarking Floating-Point Compressors

TL;DR

This paper provides a collection of real-world floating-point datasets to facilitate benchmarking of compression algorithms, addressing the need for representative data in scientific computing.

Contribution

It introduces publicly accessible datasets and sources specifically designed for evaluating floating-point data compressors.

Findings

Provides a curated list of datasets for benchmarking

Facilitates fair comparison of compression algorithms

Supports scientific computing applications

Abstract

Compression of floating-point data, both lossy and lossless, is a topic of increasing interest in scientific computing. Developing and evaluating suitable compression algorithms requires representative samples of data from real-world applications. We present a collection of publicly accessible sources for volume and time series data as well as a list of concrete datasets that form an adequate basis for compressor benchmarking.