Heavy Hitter Comparison

Running the notebook can be done as follows:

$ source MY_ENV/bin/activate 
$ python3 -m pip install pandas

Exploratory Query Cost

To produce Poplar data:

1. Run gen_params.py to produce inputs for poplar_bench_gen.py. Copy these into the poplar_bench_gen.py script.
2. Run poplar_bench_gen.py to produce poplar_bench.sh inputs. Copy these into poplar_bench.sh.
3. Clone the reference implementation, copy poplar_bench.sh into the root directory, and then run poplar_bench.sh to completion.
4. Run format_poplar_bench.py with input poplar_bench_out.txt to produce the values copied into the notebook.

To produce STAR data:

1. Clone the reference implementation.
2. Run star_bench.sh from the root directory of the repository.
3. Run format_star_bench.py with input star_bench_out.txt to produce the values copied into the notebook.