SOSD
收藏资源简介:
SOSD是一个包含8个不同数据集的基准框架,由慕尼黑工业大学创建,用于评估学习型索引结构。每个数据集包含2亿个64位无符号整数,数据来源多样,包括书籍销售、Facebook用户ID、OpenStreetMap位置等。数据集创建过程中,使用了多种算法和技术来模拟和优化数据分布。该数据集主要应用于数据库索引结构的性能评估,旨在通过比较学习型索引与传统索引的性能,推动数据库索引技术的进步。
SOSD is a benchmark framework encompassing 8 distinct datasets, developed by the Technical University of Munich for evaluating learned index structures. Each dataset consists of 200 million 64-bit unsigned integers, sourced from diverse domains including book sales records, Facebook user IDs, OpenStreetMap location data, and others. During the dataset construction phase, various algorithms and techniques were utilized to simulate and optimize the data distributions. This benchmark framework is primarily used for performance evaluation of database index structures, with the goal of advancing database indexing technology by comparing the performance of learned indexes and traditional indexes.




