The most important code used in my Masters Thesis. This code was not written to be readable or reused by people who are not Abe Wits; development speed was favored. I am considering releasing some of the tools that may be interesting for the general public; if this happens, a separate repository with the improved code + documentation will appear.

An overview of the folders in this repository:

experiment_nus
> experiments to test Non Uniform Sampling techniques for aggregation

experiment_aws
> experiments to test Heuristic Weighted Sampling

experiment_full
> experiments to test sample-Join + aggregation strategies

randomreadtimes
> tool to test read speed in memory and on disk

runtime_comparison
> tool to compare runtime of WS-sample join and US-sample join when the relations are in different storage media