Features

Features

• Synthetic Data Generation: Create datasets with specified (K,L)-sortedness, enabling controlled experiments on data with varying degrees of order.
• PostgreSQL Benchmarking Support: Evaluate PostgreSQL's performance in terms of index construction and navigation on datasets with different sortedness levels.
• Extensible Framework: While initially supporting PostgreSQL, BoDS is designed to be extensible, allowing for future integration with other database systems and indexing structures.
• Reproducible: Workloads are fully described in TOML files and can be regenerated at any point.
• Flexible Workload Descriptions: Streams can be broken up into multiple partitions, window sizes can be specified, swaps can be generated via a specified distribution, and data can be gapped.
• Explainability: The workload analyzer characterizes a workload from a datastream.
• Varying Trends: Data streams can either be semi-sorted in increasing or decreasing order.