Matlab visualization

Python visualization

Preqrequisites

• Python 3.7
• pipenv (pip install pipenv)
  • In case there is an error such as module is not callable in the install phase, downgrade pip pip install pip==18.0 and pipenv run pip install pip==18.0
• Rust's cargo if you want to run simulations. In this case, don't use repo as-is but use core-cbc: git clone core-cbc, git submodules init, cd visualization)

Install

• pipenv install

Run

• pipenv run jupyter notebook
• If the figure is not interactive, you may have to re-run all cells (manually?) to activate the interactive mode.

Run (standalone)

• Install tcl/tk
  • sudo apt-get install tk or sudo pacman -S tk
• pipenv run python visualization.py or
• pipenv run python process_metrics.py
• pipenv run python visualization_metrics.py

Scripts

• process_metrics.py Process all stats*.log files located in the parent folder. Outputs two files:

  • gen.csv which contains data points each time a validator reaches a new consensus height.
  • gen_averages.csv which is the average for all values for all validators for each run.

• aggregate.sh Aggregates all the csv files located in ./generated/backup/ that start with the prefixed passed as argument. Output files are ./gen.csv and ./gen_averages.csv.

• start_testing.sh starts multiple runs of the blockchain integration test. saves all the log files, as well as the gen.csv and gen_averages.csv. parameters:

  • prefix: prefix of the log files
  • nb_jobs: number of rust testing jobs

  example: ./start_testing.sh double_round_robin 8

• process.sh Extracts blockchain view from a single blockchain integration test run. Extracted json files are saved into ./generated/processed_states*.log

• visualization.py Shows blockchain views contained in generated/processed_states*.log

Usage

Run simulations

If you want to run simulations:

• Go to ../tests/generative_tests.rs
  • In the fn blockchain() test, in the runner.run function: Change parameters, such as height, receiver strategy, sender strategy, consensus height, number of nodes.
• In the start_testing.sh script: change the number of executions in the main function.
• Run start_testing.sh prefix core and go grab a coffee. You should now have a bunch of gen*.csv files in generated/backup/
• Run aggregate.sh prefix to aggregate all the files in generated/backup/
• You can now visualize the data using the next chapter

Use visualizations

• Launch jupyter pipenv run jupyter-notebook
  • This opens jupyter in your browser.
• Open the visualization_metrics.ipynb in jupyter
• If you want to see all the graphs with the data available on the repository:
  • Click Kernel>Restart and Run All
• If you want to show data that are just out of your simulations:
  • Change if False into if True in the second cell.
  • Run the two first cells of the notebook. (Shift-Enter)