A web application for analyzing database index selection approaches.

Based on our extensible and open-source index selection evaluation platform.

The application is built on our index selection evaluation platform, which has to be set up and run first:

1. Initialize the required git submodules

git submodule update --init --recursive

2. Install PostgreSQL, e.g., version 15

sudo apt -y install postgresql-15 postgresql-client-15 postgresql-server-dev-15

3. Set up the database user

sudo -u postgres createuser -s $(whoami);

4. Install the PostgreSQL extension HypoPG for hypothetical indexes

cd index_selection_evaluation/hypopg
make
sudo make install
cd ../..

5. Create the virtual Python environments and install the required packages.

python -m venv .pyvenv
source .pyvenv/bin/activate

pip install -r index_selection_evaluation/requirements.txt
pip install -r requirements.txt

6. Run an index selection evaluation to, then, analyze

cd index_selection_evaluation
python -m selection [CONIGURATION.JSON]

7. Run the web application

flask --app flaskr run --debug

Workloads and results to analyze can be added or edited in the HTML's dropdown and backend's benchmark configuration.