IG

Purpose

Prints unique tuples in CSV according to the Wisconsin Benchmark scalable implementation.

Relation Generation

rel_gen.py takes number of desired unique tuples as only argument, or defaults to 10, max implemented 100 million.

Note: Prints to terminal

Usage:

with headers:

python3 rel_gen.py 100000000 > someFile.txt

without headers (any second arg disables headers):

python3 rel_gen.py 100000000 n > someFile.txt

Data compresses significantly(since its mostly 'x') for transfering.

Example Data

tenktup.csv, hunktup.csv

Sample of data uploaded to bigQuery:

The code for generating the string data was adapted into Python from The Wisconsin Benchmark: Past, Present, and Future C example implementation. It can generate up to 26⁷ unique tuples.