Differential Datalog Benchmarks

This repository contains some simple benchmarks based on artificially generated datasets for the Differential Datalog engine.

Prerequisites

• Rust >= 1.50
• Python >= 3.8

How-To

1. Follow Differential Datalog's installation instructions

2. Have a look at query.dl and choose which query you would like to run.

3. Use the provided Makefile to compile the query, generate the datasets, and run the query against on of the datasets. Each dataset also has a trace of updates associated with it (see updates/), which will be fed stepwise to the query computations, and thereby showcase the incremental computation aspects.

4. You can inspect the output of the query in the respective .out files that will be generated. The benchmark results are visible in the console.

As a shorthand, you can use make all to execute the default query (computing the roots) over all datasets after you have installed Differential Datalog.