## Characterization of OutageReport for years 2019 and 2020

This repository contains two datasets extracted from OutageReport webpages in 2019 and 2020. One dataset of total reported failures and another of failure with detailed information.

The datasets are in the Parquet file format.

The Jupyter Notebook files read the datasets, clean them, and generate plots and numbers for characterization.

All the plots used in the paper can be generated by running the jupyter notebooks in order. The plots can also ve viewed in the notebooks themselves.

The easiest way to run the notebooks is to use the prebuilt docker container using the command `docker run --rm -p 8888:8888 sacheendra/or_char`.

To reproduce the graphs in the paper:
1. Start the docker container. (If you are using Docker for Mac or Windows, you might need to expose port 8888 between docker vm and host)
2. Navigate to localhost:8888 in your browser.
3. On the webpage, you should see a list of files in a pane on the left.
4. The code to reproduce the graphs in in Jupyter notebooks. Jupyter notebooks have the extension `.ipynb`.
5. The noteboks to produce the plots in the paper are numbered from 1 to 5.
6. To run a notebook, first open it by double clicking it.
7. The code in the notebook can then be run by opening the `Run` dropdown at top-left, and clicking on the `Run all cells` option.

Figure to notebook reference:
Figure 1 - Notebook 1
Table 2 - Notebook 2
Figure 2 - Notebook 3
Figure 3 - Notebook 2
Figure 4 - Experiment in Notebook 4, Plot in Notebook 5