Floating-point Compressor Benchmark (FCBench)

FCBench是由华盛顿州立大学等机构创建的数据集，包含33个来自不同领域的真实世界数据集，如科学模拟、时间序列、观测和数据库交易。这些数据集用于评估和比较13种不同的无损浮点数据压缩方法，旨在帮助研究人员选择或开发适用于集成数据库和HPC应用的压缩技术。数据集的创建过程涉及从多个来源收集和整理数据，确保数据的多样性和代表性。FCBench的应用领域广泛，包括但不限于科学计算、数据分析和存储优化，旨在解决数据存储和通信开销问题，提高数据处理性能。

FCBench is a dataset developed by Washington State University and other institutions. It includes 33 real-world datasets spanning diverse domains such as scientific simulations, time series, observational data, and database transactions. These datasets are used to evaluate and compare 13 distinct lossless floating-point data compression methods, aiming to assist researchers in selecting or developing compression technologies suitable for integrated databases and high-performance computing (HPC) applications. The creation process of FCBench involves collecting and curating data from multiple sources to ensure its diversity and representativeness. FCBench has a wide range of application scenarios, including but not limited to scientific computing, data analysis, and storage optimization, and it is intended to address issues related to data storage and communication overhead, thereby enhancing data processing performance.